text
stringlengths 66
377k
|
|---|
First reported use of real-time intraoperative computed tomography angiography image registration using the Machine-vision Image Guided Surgery system: illustrative case
BACKGROUND Vertebral artery injury is a devastating potential complication of C1-2 posterior fusion. Intraoperative navigation can reduce the risk of neurovascular complications and improve screw placement accuracy. However, the use of intraoperative computed tomography (CT) increases radiation exposure and operative time, and it is unable to image vascular structures. The Machine-vision Image Guided Surgery (MvIGS) system uses optical topographic imaging and machine vision software to rapidly register using preoperative imaging. The authors presented the first report of intraoperative navigation with MvIGS registered using a preoperative CT angiogram (CTA) during C1-2 posterior fusion.OBSERVATIONS MvIGS can register in seconds, minimizing operative time with no additional radiation exposure. Furthermore, surgeons can better adjust for abnormal vertebral artery anatomy and increase procedure safety.LESSONS CTA-guided navigation generated a three-dimensional reconstruction of cervical spine anatomy that assisted surgeons during the procedure. Although further study is needed, the use of intraoperative MvIGS may reduce the risk of vertebral artery injury during C1-2 posterior fusion.https://thejns.org/doi/abs/10.3171/CASE2125
C1-2 fusion is a challenging procedure because of complex neurovascular anatomy that increases surgical risks and necessitates highly accurate screw positioning. Although a rare occurrence, vertebral artery injury during cervical spine surgery can cause catastrophic bleeding, permanent neurological impairment, stroke, and/ or death. [bib_ref] Analysis of anatomical variations of bone and vascular structures around the posterior..., Hong [/bib_ref] [bib_ref] Risk of vertebral artery injury: comparison between C1-C2 transarticular and C2 pedicle..., Yeom [/bib_ref] Reported complication rates vary based on surgical approach, with C1-2 posterior fusion carrying the highest risk of vertebral artery injury, ranging from 4.1% to 8.2% of cases. [bib_ref] Vertebral artery injury in C1-2 transarticular screw fixation: results of a survey..., Wright [/bib_ref] [bib_ref] Radiological and anatomical evaluation of the atlantoaxial transarticular screw fixation technique, Madawi [/bib_ref] [bib_ref] Iatrogenic vertebral artery injury during anterior cervical spine surgery, Burke [/bib_ref] [bib_ref] Vertebral artery complications in anterior approaches to the cervical spine: report of..., Daentzer [/bib_ref] Advanced imaging is an essential part of preoperative assessment, and a computed tomography angiogram (CTA) is often obtained preoperatively to carefully delineate vascular anatomy and provide guidance on safe screw placement. Use of intraoperative fluoroscopy or computer-assisted navigation during surgery may further reduce surgical complication risk and improve screw accuracy. [bib_ref] Accuracy of pedicle screw insertion among 3 image-guided navigation systems: systematic review..., Du [/bib_ref] [bib_ref] The accuracy of pedicle screw placement using intraoperative image guidance systems, Mason [/bib_ref] The Machine-vision Image Guided Surgery (MvIGS) system (7D Surgical, Inc.) is an intraoperative spinal navigation system that uses optical topographic imaging. Unlike other navigation systems that rely on intraoperative fluoroscopy or CT scans, MvIGS uses nonionizing structured light to obtain a three-dimensional (3D) surface scan of patient anatomy that is automatically registered to the patient's preoperative CT scan within seconds. 9,10 This technology was initially developed to improve surgical workflow and allow intraoperative CT-guided navigation without the need for intraoperative scanning time and without intraoperative surgeon, staff, and patient radiation exposure. However, as this technology is used in more spinal procedures, additional advantages of MvIGS become apparent, such as the ability to navigate intraoperatively with a CTA instead of a standard CT scan. This advantage allows simultaneous visualization of bony and vascular anatomy and the critical spatial relationships between these structures. It also enables visualization of the extraosseous course of the vertebral artery, which is particularly useful in posterior C1-2 fusion procedures.
To the best of our knowledge, 7D's MvIGS technology is the only navigation system that allows for real-time intraoperative navigation guidance with a CTA, and this is the first reported case using the technology in a C1-2 posterior fusion.
## Illustrative case history and presentation
An 86-year-old woman with severe osteoporosis (T-score −6.3), hypertension, and depression suffered a type 2 displaced odontoid fracture after a ground-level fall that was conservatively managed with an Aspen collar at an outside institution. She presented to our clinic several months after the fall with persistent axial neck pain, intermittent numbness in both hands, and unsteady gait. Her neurological exam was normal without evidence of myelopathy. Magnetic resonance imaging of the patient's cervical spine demonstrated a type 2 odontoid fracture with posterior displacement of the fractured dens and evidence of nonunion [fig_ref] FIG. 1: Preoperative radiograph [/fig_ref]. Given the failure of bony fusion with conservative therapy and the encroachment of the displaced odontoid fracture into the canal, surgical intervention was recommended.
## Operative technique
The patient was positioned prone on a Jackson table with her head fixed in the radiolucent Mayfield frame. After subperiosteal exposure of the dorsal bony anatomy of the atlas and axis (with removal of all soft tissue), the MvIGS system was brought into the field, and the clamp was placed on the C2 spinous process. Flash registration was performed, matching 1924 points from the C1 posterior arch and 1877 points from the C2 lamina and pars onto the preoperative CTA , D, and E). The entire registration process took 171 seconds (<3 min) and allowed us to navigate our screw placement based on 3D reconstructions of the cervical spine that included the vascular anatomy . First, we used the navigated probe [fig_ref] FIG. 3: Navigated screw placement [/fig_ref] to determine the screw trajectory, which was started with the high-speed air drill. Next, we used the navigated drill guide [fig_ref] FIG. 3: Navigated screw placement [/fig_ref] to drill the screw trajectory and the navigated tap [fig_ref] FIG. 3: Navigated screw placement [/fig_ref] to tap the trajectory. The screws were placed using the virtual K-wire feature of the MvIGS system [fig_ref] FIG. 3: Navigated screw placement [/fig_ref]. All screws accurately followed the trajectory mapped by the MvIGS system, as indicated in the figure with the planned trajectories superimposed on the final screw placement seen in the intraoperative O-arm spin . A fibular strut graft was also placed in the C1-2 space and secured via a wiring technique to encourage bony fusion. An intraoperative CT scan was obtained to confirm accurate screw placement, per the surgeon's standard protocol.
## Postoperative course
The patient remained neurologically intact and was discharged to a skilled nursing facility after an uncomplicated hospital course. At her 6-week and 3-month postoperative visits, the patient reported improving incisional neck pain and walked well with a walker. Postoperative radiographs demonstrated excellent hardware position .
# Discussion
## Observations
Atlantoaxial fixation is challenging because of the complex regional anatomy, the small size of the C1 lateral masses and C2 pedicles, and the proximity of important neurovascular structures, including the vertebral arteries. [bib_ref] A review of complications associated with craniocervical fusion surgery, Lall [/bib_ref] [bib_ref] The clinical risk of vertebral artery injury from cervical pedicle screws inserted..., Neo [/bib_ref] The incidence of anomalous vertebral arteries ranges from 2.7% in cadaveric studies to 5.4% in imaging studies.Although the incidence of vertebral artery injury is low, such injuries can be devastating and lead to permanent neurological injury, stroke, or death in 10% to 20% of cases. [bib_ref] Vertebral artery injuries in cervical spine surgery, Lunardini [/bib_ref] [bib_ref] Imaging for blunt carotid and vertebral artery injuries, Burlew [/bib_ref] Although previous navigation systems enabled the use of a preoperative CT scan for intraoperative real-time navigation, [bib_ref] Screw placement accuracy and outcomes following O-arm-navigated atlantoaxial fusion: a feasibility study, Smith [/bib_ref] angiograms had not yet been used for this purpose. To the best of our knowledge, this study is the first to demonstrate the feasibility of real-time intraoperative navigation with MvIGS using a preoperative CTA.
As discussed previously, MvIGS system technology registers onto a preoperative CT scan or CTA in the span of seconds to a few minutes. [bib_ref] Machine vision augmented reality for pedicle screw insertion during spine surgery, Nguyen [/bib_ref] In addition, if the patient or the reference clamp moves during the procedure, the system can be readjusted rapidly with a repeat Flash Fix registration that takes seconds, although this feature was not used in our case. For C1-2 posterior fusions in particular, moving the spinous process clamp can be helpful between screw placements when the clamp is in the way of a specific screw trajectory. This technique has been used by the senior author (C.C.Z.) in other cases of C1-2 instrumentation. Also of note, the MvIGS system is a navigation platform only and can be used with spine implants from any company.
## Lessons
Yamazaki and colleagues have demonstrated the value of preoperative 3D CTA reconstructions for identifying abnormal vertebral artery anatomy. [bib_ref] Anomalous vertebral arteries in the extra-and intraosseous regions of the craniovertebral junction..., Yamazaki [/bib_ref] Although CT-guided navigation is becoming the standard of care in spine surgery,with improved screw accuracy and decreased fluoroscopic time and blood loss compared to those with traditional fluoroscopic and free-hand techniques, [bib_ref] Comparison of isocentric C-arm 3-dimensional navigation and conventional fluoroscopy for C1 lateral..., Yang [/bib_ref] traditional spinal navigation systems are unable to visualize vascular anatomy. Hsu et al. found that 50% of cases of vertebral artery injury revealed anomalous vessel anatomy during postoperative review, [bib_ref] Epidemiology and outcomes of vertebral artery injury in 16 582 cervical spine..., Hsu [/bib_ref] which highlights the importance of preoperative and, potentially, intraoperative visualization of vertebral vasculature. MvIGS offers the additional benefit of intraoperative navigation using a 3D CTA reconstruction, which may enhance surgeon screw placement accuracy and avoid vascular complications. Although navigation systems, including those by 7D, require additional upfront setup time, operating room efficiency may improve in the long term as surgeons and their operative teams overcome the learning curve. [bib_ref] Effect of intraoperative navigation on operative time in 1-level lumbar fusion surgery, Khanna [/bib_ref] We present a case of CTA-guided navigation during C1-2 posterior fusion for type 2 odontoid fracture using MvIGS. The MvIGS system minimizes workflow disruption and does not increase radiation exposure. 3D reconstruction of the regional bony and vascular anatomy assists surgeons during screw placement. CTA-guided navigation offers promise in assisting surgeons during complex craniocervical cases.
[fig] FIG. 1: Preoperative radiograph (A), CT scan (B), and T2-weighted magnetic resonance image (C) of type 2 odontoid fracture with posterior displacement of dens. FIG. 2. MvIGS system. Bony landmarks on C1 and C2 were selected (A and D), and registration of C1 (B) and C2 (E) vertebrae was performed. A navigated probe was used to guide screw trajectory (C and F) on a 3D reconstruction of vertebral artery anatomy (G). [/fig]
[fig] FIG. 3: Navigated screw placement. A navigated cervical pedicle probe (A), drill guide with universal tracker (B), and tap with universal tracker (C) were used to drill and tap screw trajectory. The virtual K-wire feature (D and E) was used to guide screw placement. FIG. 4. Screw trajectory alignment. Pedicle screws were placed into the C1 (A and B) and C2 (C and D) vertebrae along the trajectories mapped by the MvIGS system. [/fig]
|
PALM: A Paralleled and Integrated Framework for Phylogenetic Inference with Automatic Likelihood Model Selectors
Background: Selecting an appropriate substitution model and deriving a tree topology for a given sequence set are essential in phylogenetic analysis. However, such time consuming, computationally intensive tasks rely on knowledge of substitution model theories and related expertise to run through all possible combinations of several separate programs. To ensure a thorough and efficient analysis and avert tedious manipulations of various programs, this work presents an intuitive framework, the phylogenetic reconstruction with automatic likelihood model selectors (PALM), with convincing, updated algorithms and a best-fit model selection mechanism for seamless phylogenetic analysis.Methodology: As an integrated framework of ClustalW, PhyML, MODELTEST, ProtTest, and several in-house programs, PALM evaluates the fitness of 56 substitution models for nucleotide sequences and 112 substitution models for protein sequences with scores in various criteria. The input for PALM can be either sequences in FASTA format or a sequence alignment file in PHYLIP format. To accelerate the computing of maximum likelihood and bootstrapping, this work integrates MPICH2/ PhyML, PalmMonitor and Palm job controller across several machines with multiple processors and adopts the task parallelism approach. Moreover, an intuitive and interactive web component, PalmTree, is developed for displaying and operating the output tree with options of tree rooting, branches swapping, viewing the branch length values, and viewing bootstrapping score, as well as removing nodes to restart analysis iteratively.Significance: The workflow of PALM is straightforward and coherent. Via a succinct, user-friendly interface, researchers unfamiliar with phylogenetic analysis can easily use this server to submit sequences, retrieve the output, and re-submit a job based on a previous result if some sequences are to be deleted or added for phylogenetic reconstruction. PALM results in an inference of phylogenetic relationship not only by vanquishing the computation difficulty of ML methods but also providing statistic methods for model selection and bootstrapping. The proposed approach can reduce calculation time, which is particularly relevant when querying a large data set. PALM can be accessed online at http://palm.iis.
# Introduction
Advances in molecular biology and bioinformatics have enabled researchers to obtain gene sequences by experimental procedures, as well as by sequence searching approaches. The feasibility of utilizing sequence features that can be viewed as evolutionary changes in the sequence occurring in the operating taxonomic units has received considerable interest. A molecular phylogenetic analysis procedure based on sequence features is initiated by gathering a set of sequences derived from a common origin. Corresponding residues among DNA/protein sequences are then defined by multiple sequence alignment (MSA). Next, the relatedness among the input sequences is estimated using the phylogenetic inference method with a suitable substitution model. Finally, the representative tree(s) of the phylogenetic relationship is constructed and may be presented graphically with statistical confidence of the branching topology. Significant advances in theoretical and mathematical implementation of phylogenetic methodology have been made in recent decades. These methodological advances provide an unprecedented and often bewildering set of choices on methodological issues [bib_ref] Troubleshooting molecular phylogenetic analyses, Sanderson [/bib_ref]. Researchers unfamiliar with phylogenetic analysis are perplexed by selecting and installing programs, transferring files between programs, setting parameters for the complete process, running iterations with various combinations of methods and parameters [bib_ref] Choosing a Method for Phylogenetic Prediction, Mount [/bib_ref] , as well as recalling all tools and parameter-related choices made during analysis.
Although the relationship among an interesting sequence set from an evolutional aspect has received considerable attention, performing adequate phylogenetic analysis with a sound theoretical foundation is quite difficult.
Among the three major categories of phylogenetic inference methods, distance, maximum parsimony (MP), and maximum likelihood (ML), ML methods are especially useful for sequence sets with varying extents of sequence diversity [bib_ref] Choosing a Method for Phylogenetic Prediction, Mount [/bib_ref] [bib_ref] The troubled growth of statistical phylogenetics, Felsenstein [/bib_ref]. Based on the theory of ML methods, the likelihood of a series of residue substitution events is estimated and, then, the most probable tree topology from all possible ones that represents the evolutionary history of a given sequence set can be inferred. Various residue substitution models that describe the probability of replacing one residue with another have been derived from either statistical analysis involving conserved sequence blocks or molecular evolution theories [bib_ref] Protein evolution with dependence among codons due to tertiary structure, Robinson [/bib_ref] [bib_ref] Protein evolution constraints and model-based techniques to study them, Thorne [/bib_ref] , and can be further refined for parameters involving sites under the selection force or for varying substitution rates among relevant sites. Likelihood-based approaches are robust for phylogenetic inference. Although obtaining a best-fit ML tree with an increasing number of sequences and sequence length is computationally intractable (NP-hard), a ML-based method can provide statistical comparability of the fitness of substitution models [bib_ref] Maximum likelihood of evolutionary trees: hardness and approximation, Chor [/bib_ref] [bib_ref] Molecular phylogenetics: state-of-the-art methods for looking into the past, Whelan [/bib_ref].
There are several ML-based phylogenetic inference programs available, including PhyML [bib_ref] A simple, fast, and accurate algorithm to estimate large phylogenies by maximum..., Guindon [/bib_ref] [bib_ref] Online-a web server for fast maximum likelihood-based phylogenetic inference, Guindon [/bib_ref] , PAML [bib_ref] PAML: a program package for phylogenetic analysis by maximum likelihood, Yang [/bib_ref] [bib_ref] PAML 4: phylogenetic analysis by maximum likelihood, Yang [/bib_ref] , Multiphyl [bib_ref] MultiPhyl: a high-throughput phylogenomics webserver using distributed computing, Keane [/bib_ref] , Phylip package [bib_ref] Inferring phylogenies from protein sequences by parsimony, distance, and likelihood methods, Felsenstein [/bib_ref] and IQPNNI [bib_ref] IQPNNI: moving fast through tree space and stopping in time, Vinh Le [/bib_ref]. Owing to a heavy computational load, several attempts have been made to accelerate the estimation of ML by refining algorithms and applying parallel computing, e.g., PAML [bib_ref] PAML: a program package for phylogenetic analysis by maximum likelihood, Yang [/bib_ref] [bib_ref] PAML 4: phylogenetic analysis by maximum likelihood, Yang [/bib_ref] , RAxML [bib_ref] RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees, Stamatakis [/bib_ref] , GARLI [bib_ref] Genetic algorithms and parallel processing in maximum-likelihood phylogeny inference, Brauer [/bib_ref] , mpi-CH2/PhyML [bib_ref] Online-a web server for fast maximum likelihood-based phylogenetic inference, Guindon [/bib_ref] and pIQPNNI [bib_ref] pIQPNNI: parallel reconstruction of large maximum likelihood phylogenies, Minh [/bib_ref]. These programs are performed in command-line mode or in a graphical-user interface. Some of the phylogenetic servers, such as PhyML web server [bib_ref] Online-a web server for fast maximum likelihood-based phylogenetic inference, Guindon [/bib_ref] , Multiphyl web server [bib_ref] MultiPhyl: a high-throughput phylogenomics webserver using distributed computing, Keane [/bib_ref] , phylogeny.fr [bib_ref] Phylogeny.fr: robust phylogenetic analysis for the non-specialist, Dereeper [/bib_ref] and Phylemon [bib_ref] Phylemon: a suite of web tools for molecular evolution, phylogenetics and phylogenomics, Tarraga [/bib_ref] , are available, providing computational power and a user-friendly interface. Users should normally be aware of phylogenetic analysis, and then achieve a reasonable outcome from these web applications.
How to select an optimal substitution model for ML estimation is addressed in [bib_ref] Selecting the best-fit model of nucleotide substitution, Posada [/bib_ref]. The implementations of best model selection procedures were proposed for both protein sequences [bib_ref] ProtTest: selection of best-fit models of protein evolution, Abascal [/bib_ref] and nucleotide sequences [bib_ref] MODELTEST: testing the model of DNA substitution, Posada [/bib_ref]. ProtTest [bib_ref] ProtTest: selection of best-fit models of protein evolution, Abascal [/bib_ref] is a model selection scheme for protein sequences based on score files generated by PhyML. The fitness of amino acid substitution models is estimated using three scores, i.e. Akaike information criterion (AIC), corrected AIC (AICc) and Bayesian Information Criterion (BIC), as well as the maximum likelihood of each model. Modeltest [bib_ref] MODELTEST: testing the model of DNA substitution, Posada [/bib_ref] applies AIC, AICc, BIC, and hierarchical likelihood ratio tests (hLRTs), to estimate the model fitness from the score file of DNA substitution models generated by PAUP [bib_ref] Inferring evolutionary trees with PAUP*, Wilgenbusch [/bib_ref]. Currently, MOD-ELTEST has been superseded by jModeltest [bib_ref] jModelTest: phylogenetic model averaging, Posada [/bib_ref] , an integrated framework using PhyML [bib_ref] A simple, fast, and accurate algorithm to estimate large phylogenies by maximum..., Guindon [/bib_ref] computational procedures to provide the nucleotide substitution model selection. jModeltest provides two additional evaluation criteria, i.e. dynamical likelihood ratio tests (dLRT) and a decision theory method (DT), as estimates of model selection uncertainty, parameter importance and modelaveraged parameter estimates, including model-averaged phylogenies. FINDMODELimplemented the idea of MODELTEST for testing the best fit model for the input alignment of nucleotide sequences. FINDMODEL provides options of selecting the model set and tree methods for constructing the initial tree. The best substitution model from 28 models is determined by the lowest AIC of the optimized ML tree topology, as estimated by baseml (PAML) and MODELTEST. The model estimators described above take aligned sequences as input and select the most appropriate model based on the probability (the likelihood) of the ML tree. Once the model has been set, a ML-based phylogenetic inference program can be applied for bootstrapping and the possible phylogenetic relationship presented in a tree topology obtained as well. The entire process, from the model selection to bootstrapping, is time consuming and computationally intractable, especially when handling large data sets.
To perform comprehensive phylogenetic analysis with a model selection mechanism without monotonous manipulations of data input/output, this work describes the design of an intuitive framework for phylogenetic reconstruction with automatic likelihood model selectors (PALM). By using the proposed integrated framework of ClustalW, PhyML, MODELTEST, ProtTest, and several in-house programs (PalmDaemon, PalmMonitor, PalmTree, and Palm job controller), the fitness of 56 substitution models is evaluated for nucleotide sequences and 112 substitution models for protein sequences with scores in various criteria (i.e. AIC, AICc, BIC and hLRT). Via the parallel computing strategy, the calculation time of this ML-based phylogenetic reconstruction tool is substantially reduced. All parameters used and outputs generated by each program can be accessed online. Moreover, resubmitting a new job from previous graphic results to remove rogue taxa and add new ones is an effortless task when using PalmTree.
# Methods
## Structure of palm system
This work adopts the maximum likelihood (ML) method, which facilitates the application of mathematical models that incorporate the knowledge of typical patterns of sequence evolution, resulting in more powerful phylogenetic inferences [bib_ref] Molecular phylogenetics: state-of-the-art methods for looking into the past, Whelan [/bib_ref]. To ensure a stable system with satisfactory performance, PALM is constructed on a platform with several symmetrical multi-processor (SMP) servers equipped with quad CPUs (700MHz Intel Xeon) and 8 GB RAM. Additionally, the LAMP structure (Ubuntu, version 8.04; Apache, version 2.04; Postgresql, version 8.3.7; PHP, version 5.1.0) and a MS-Window 2003 server are adopted to take the computation load and streamline the workflow control (PalmDaemon).
PALM is designed for biologists without prerequisite computer skills to run a complex phylogenetic analysis process via a succinct, user-friendly interface. Achieving this objectives involves integrating several well-established programs (i.e. readseq
## Palmdaemon
Based on the design of PALM, a series of successive calculation steps is transformed into an integrated service. PalmDaemon is responsible for transferring data between programs and agglutinating several prestigious programs and our in-house programs. Following selection of the optimum model, the alignment, parameters for model choice, and bootstrap setting are submitted to Palm job controller for parallel computing on bootstrapping. To easily retrieve the output results, PalmDaemon sends e-mails notifying job acceptance and job completion to users. Following the job ID link in the notification mail, users can trace back to their job results. Query parameters are stored and used if a resubmission job is initiated from the PalmTree Viewer (as described below). Additionally, to avert long sequence identifier (ID) truncation caused by alignment, the sequence IDs (up to 100 characters) are converted into internal running IDs by PalmDaemon and are restored in the final outputs. Therefore, meaningful sequence IDs encoded with a message such as the sampling date and species name are allowed, subsequently making the result comprehensible.
PalmMonitor: Multi-Thread Processing PALM runs through tree-building processes of all available models and then identifies the most appropriate model in terms of the tree with the best likelihood -derived score by the model. This time consuming task is accelerated by implementing a multithread dispatch program in Java, PalmMonitor. PalmMonitor is used to divide a single sequential task of running 56 nucleic acids substitution models and 112 protein substitution models into parallel ones and, then, merges the results when all tasks are completed. Doing so reduces a considerable amount of computational time, i.e. a six-threading run reduces the computing time from eight hours to eighty minutes for a query of 20 sequences with 5000 residues in length.
## Palmtree: an interactive tree viewer and a job resubmission gateway
PalmTree is a versatile tool composed of C, php and Java languages for viewing the tree topology and restarting analysis. PalmTree parses the context of the Newick format tree file and converts the text layout into a graph. In this work, the tree topology is provided by one of three options, i.e. the original tree plot, unrooted tree, and pseudo-rooted tree by the mid-point method [bib_ref] An empirical test of the midpoint rooting method, Hess [/bib_ref]. The mid-point method is adopted as the default presentation of tree topology; the mid-point of the longest node-tonode path is set as an artificial rooting node and a pseudo-rooted tree in a balanced structure is displayed. The branch length and the bootstrapping value (in percentage) in the output tree file are optional display parameters. Additionally, the tree branch edge can be broadened according to the bootstrapping value, thus providing a direct impression on the confidence on the branching pattern. User can click on the branching point to perform the branch swapping without altering the tree topology. Notably, any leaf node, i.e. a sequence in the original query, can be removed from the tree effortlessly with a mouse click; meanwhile, undeleted sequences can be re-submitted to PALM for analysis. Additionally, new sequences can be added through the job submission form for reinitiating the analysis.
## Palm job controller: a paralleled task controller for bootstrapping
An attempt is made to enhance the performance of PALM by conducting distributed computing in a remote python call (RPyC, http://rpyc.wikidot.com/), a transparent and symmetrical python library for remote procedure calls, clustering and distributedcomputing, with mpi-PhyML. Based on this framework, multiple submissions can simultaneously be processed dynamically to overcome the physical boundaries between processers and computers. This feature implies that PALM can modify the thread from one to many in order to avoid a computationally burdensome task from occupying the entire system.
# Results
## Palm workflow
PALM analysis begins by submitting a selected sequence set or pre-aligned sequences. Submitted nucleic acid sequences or protein sequences in FASTA format are computed for the output alignment in PHYLIP format using ClustalW; a query initiated in a pre-aligned file in PHYLIP format by other sequence align tools disregards this procedure. The alignment is then transferred to an automatic model selecting process in PalmDaemon and PalmMonitor by routing through PhyML/MODELTEST for DNA and PhyML/ProtTest for protein sequences, respectively. The fitness among 56 substitution models (JC69, K80, F81, HKY, TrN, TrNef, K3P, K3Puf, TIM, TIMef, TVM, TVMef, SYM, and GTR with parameter G, I) for nucleotide sequences and 112 models (LG, DCMut, JTT, MtREV, MtMam, MtArt, Dayhoff, WAG, RtREV, CpREV, Blosum62, VT, HIVb, and HIVw with parameter I, G, F) for protein sequences is estimated in PALM, and the optimal model are used to infer the phylogeny. The models denoted by +I imply that a fraction of data set is assumed to be invariable, where +G is considered to categorize the change of substitution rates among sites in discrete gamma distribution. Besides, +F implies that the equilibrium base frequencies in the sequences are estimated by observing the occurrence in the data.
Once the maximum likelihood for all available models is estimated and merged into a file by PalmDaemon, statistical information of the parameters can be accessed to evaluate the fitness of a model for a given alignment. In this work, three criteria are provided, i.e. Akaike Information Criterion (AIC, and the derived AICc and AICw), Bayesian Information Criterion (BIC) and the maximum likelihood value (LnL), as customized options for ranking the fitness of models. Finally, the top ranked model is selected for the phylogenetic inference with iterations via PhyML, and a tree topology with a bootstrap value is generated. The user is notified via e-mail with a link to the PALM result page after the task is completed [fig_ref] Figure 1: Infrastructure and workflow of PALM [/fig_ref]. [fig_ref] Figure 2: PALM Output [/fig_ref] summarizes the results of PALM, as categorized in five parts. Job parameters refer to submission-related information, including a job ID, sequence type, bootstrap number, model selection criterion and number of substitution rate categories. A tree image area is the graph output from PalmTree. Information for the best model describes the summary of the model being selected for the phylogenetic inference. An expendable, information for all model tables displays the details of the maximum likelihood for models sorted by the ranking option set for model selection. All output files generated during the reconstruction process are available in the download area. Moreover, PALM provides a mechanism to reinitiate the analysis for a user to add and remove sequences on each experiment via PalmTree (tree image area), as mentioned earlier. This mechanism allows users to identify inappropriate/irrelevant sequences and delete sequences in a visualized environment.
## Palm output
# Discussion
PALM is an integrated framework of ClustalW, PhyML, MODELTEST, ProtTest and several in-house programs to evaluate the fitness of 56 substitution models for nucleotide sequences and 112 substitution models for protein sequences with scores in various criteria. It is especially useful for biologists to perform phylogenetic analyses without prerequisite computer skills. Users are free of tedious tasks of sequence/file format conversion. Problems incurred by sequence alignment on long descriptions as sequence identifiers are also resolved. Hence, the sequence IDs can be displayed correctly for telling biological truths in both the tree topology and in the newick format tree file. In contrast with three other renowned phylogenetic web servers, Phylogeny.fr [bib_ref] Phylogeny.fr: robust phylogenetic analysis for the non-specialist, Dereeper [/bib_ref] , Phylemon [bib_ref] Phylemon: a suite of web tools for molecular evolution, phylogenetics and phylogenomics, Tarraga [/bib_ref] and Multiphyl [bib_ref] MultiPhyl: a high-throughput phylogenomics webserver using distributed computing, Keane [/bib_ref] , the entire workflow of PALM is straightforward and coherent tightly from the job submission to the output retrieving with even more efficiency and more available substitution models. Unlike Phylogeny.fr, PALM does not allow users to define their choice stepwise. For users who prefer to set all their choices, a complementary action to PALM can be performed in our previous work, POWER (PhylOgenetic WEb Repeater, http://power.nhri. org.tw) [bib_ref] POWER: PhylOgenetic WEb Repeater-an integrated and user-optimized framework for biomolecular phylogenetic analysis, Lin [/bib_ref] which is designed for running the phylo-genetic inference based on Phylip package (http://evolution.gs.washington. edu/phylip.html) for specific methods of phylogenetic inference (mainly in the distance method, maximum parsimony, and some maximum likelihood method). Otherwise, users may execute their request on the expert mode of phylogeny.fr or phylemon for ML methods with phylogeny knowledge.
There are several MSA tools available, including ClustalW and ClustalX [bib_ref] Multiple sequence alignment using ClustalW and ClustalX, Thompson [/bib_ref] , MUSCLE [bib_ref] MUSCLE: a multiple sequence alignment method with reduced time and space complexity, Edgar [/bib_ref] MAFFT [bib_ref] Recent developments in the MAFFT multiple sequence alignment program, Katoh [/bib_ref] , DIALIGN [bib_ref] DIALIGN: multiple DNA and protein sequence alignment at BiBiServ, Morgenstern [/bib_ref] [bib_ref] DIALIGN-TX: greedy and progressive approaches for segment-based multiple sequence alignment, Subramanian [/bib_ref] , and T-coffee [bib_ref] T-Coffee: A novel method for fast and accurate multiple sequence alignment, Notredame [/bib_ref]. Essoussi et al. compared several alignment tools, indicating that no single tool can consistently outperform other ones by the estimation of quality alignment [bib_ref] A comparison of MSA tools, Essoussi [/bib_ref]. This work thus incorporates ClustalW as the built-in alignment tool of PALM owing to its stable performance and popularity. Meanwhile, prealigned sequences in PHYLIP format can be accepted by PALM as input to fulfill user requests with respect to alignment from various perspectives.
As mentioned earlier, a ML-based method is computationally difficult. After the number of input sequences, or the sequence length, or the number of available models is increased, the computing task of ML burdens the system. For instance, jModeltest [bib_ref] jModelTest: phylogenetic model averaging, Posada [/bib_ref] was designed to estimate the optimal model starting from aligned sequences. A large dataset ran in jModeltest is time consuming (e.g., 4 hours for 24 sequences on an average of 4600 bps). Smoothly implementing the system for research purposes necessitates setting PALM in a reasonable running time for a job query. For accelerating ML calculations, this work presents a novel workflow framework by integrating PhyML, MODELTEST, PalmMonitor and the Palm job controller to conquer by dividing successive jobs into several parallel processes to fully utilize most of the multi-core CPU resources. This multi-thread strategy accelerates ML calculations during the model selection stage and, in doing so, significantly reduces time consumption to about one sixth of the single, non-parallel process in the currently designated servers. The parallel computing strategy is also applied during the bootstrapping stage (mpi-PhyML/Palm job controller). However, a long running time query is possibly owing to the poor alignment of the input sequences. Users may examine the alignment of the sequences and the output tree topology, attempt to identify irrelevant sequences from the dataset and, then, resubmit the job.
The complete phylogenetic analysis workflow of PALM is a heavy computational and time-consuming task. However, we recommend incorporating contemporary models that may provide more sophisticated views on residue substitution during evolution processes. Other advanced algorithms for ML inference, e.g., DPRml [bib_ref] DPRml: distributed phylogeny reconstruction by maximum likelihood, Keane [/bib_ref] , MrBayes [bib_ref] MrBayes 3: Bayesian phylogenetic inference under mixed models, Ronquist [/bib_ref] and RAxML [bib_ref] RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees, Stamatakis [/bib_ref] , can be integrated into PALM to provide Bayesian estimates of phylogeny and handle large phylogenetic trees in the future. Refining the entire pipeline for parallel computing is currently underway in our laboratory. Plans are also underway to introduce additional high performance computing technologies, e.g., distributed computing and cloud computing, in order to mitigate the computational load of sequence alignment, likelihood estimation and bootstrapping, which is heavy along with the large amount of input data.
[fig] Figure 1: Infrastructure and workflow of PALM. doi:10.1371/journal.pone.0008116.g001 [/fig]
[fig] Figure 2: PALM Output. The result page consists of five parts. A). Job parameters. The job ID and user-defined parameters in the submission are included. B). Tree topology drawn by PalmTree, an interactive topology viewer with displaying options of bootstrapping value and branch length. A mouse click on a branching point can make the sub-tree flip; a click on the end node (round, with sequence ID) removes the sequence from the submitted data set before reinitiating an analysis procedure. C). Information about the best model selected by PalmDaemon. D). Statistics on all models calculated by PalmDaemon. E). Download area for those files generated from the entire PALM process. doi:10.1371/journal.pone.0008116.g002 [/fig]
|
lon Deletion Impairs Persister Cell Resuscitation in Escherichia coli
Bacterial persisters are nongrowing cells highly tolerant to bactericidal antibiotics. However, this tolerance is reversible and not mediated by heritable genetic changes. Lon, an ATP-dependent protease, has repeatedly been shown to play a critical role in fluoroquinolone persistence in Escherichia coli. Although lon deletion (Dlon) is thought to eliminate persister cells via accumulation of the cell division inhibitor protein SulA, the exact mechanism underlying this phenomenon is not yet elucidated. Here, we show that Lon is an important regulatory protein for the resuscitation of the fluoroquinolone persisters in E. coli, and lon deletion impairs the ability of persister cells to form colonies during recovery through a sulAand ftsZ-dependent mechanism. Notably, this observed "viable but nonculturable" state of antibiotic-tolerant Dlon cells is transient, as environmental conditions, such as starvation, can restore their culturability. Our data further indicate that starvation-induced SulA degradation or expression of Lon during recovery facilitates Z-ring formation in Dlon persisters, and Z-ring architecture is important for persister resuscitation in both wild-type and Dlon strains. Our in-depth image analysis clearly shows that the ratio of cell length to number of FtsZ rings for each intact ofloxacin-treated cell predicts the probability of resuscitation and, hence, can be used as a potential biomarker for persisters. IMPORTANCE The ATP-dependent Lon protease is one of the most studied bacterial proteases. Although deletion of lon has been frequently shown to reduce fluoroquinolone persistence, the proposed mechanisms underlying this phenomenon are highly controversial. Here, we have shown that lon deletion in Escherichia coli impairs the ability of persister cells to form colonies during recovery and that this reduction of persister levels in londeficient cells can be transient. We also found that altered Z-ring architecture is a key biomarker in both wild-type and lon-deficient persister cells transitioning to a normal cell state. Collectively, our findings highlight the importance of differentiating persister formation mechanisms from resuscitation mechanisms and underscore the critical role of the nonculturable cell state in antibiotic tolerance.
frequently shown to reduce fluoroquinolone persistence [bib_ref] Tolerance of Escherichia coli to fluoroquinolone antibiotics depends on specific components of..., Theodore [/bib_ref] , suggesting that this protein may be an attractive target for small molecular inhibitors [bib_ref] Leveraging peptide substrate libraries to design inhibitors of bacterial lon protease, Babin [/bib_ref]. However, the proposed mechanisms underlying the lon-dependent persistence state are highly controversial [bib_ref] Persistence increases in the absence of the alarmone guanosine tetraphosphate by reducing..., Chowdhury [/bib_ref] [bib_ref] Commentary: what is the link between stringent response, endoribonuclease encoding type II..., Van Melderen [/bib_ref]. Lon was initially thought to induce persistence by degrading antitoxin molecules through a ppGpp/polyphosphate-dependent mechanism, but this model is no longer supported, as the reported observations were due to artifacts from a notorious laboratory contaminant. Another suggested mechanism for the role of Lon in bacterial persistence depends on the cell division inhibitor protein SulA and is based on the observation that deletion of sulA restores fluoroquinolone persister levels in lon-deficient strains [bib_ref] Highly contingent phenotypes of lon protease deficiency in Escherichia coli upon antibiotic..., Matange [/bib_ref]. In this model, accumulation of SulA in the absence of Lon should inhibit FtsZ-dependent ring formation in fluoroquinolonetreated persister cells [bib_ref] Highly contingent phenotypes of lon protease deficiency in Escherichia coli upon antibiotic..., Matange [/bib_ref] [bib_ref] Cell division inhibitors SulA and MinCD prevent formation of the FtsZ ring, Bi [/bib_ref] [bib_ref] Inhibition of FtsZ polymerization by SulA, an inhibitor of septation in Escherichia..., Mukherjee [/bib_ref] , thus impairing persister cell resuscitation and colony formation. Notably, this hypothesis, which remains unverified, may contradict the notion of persister cell dormancy. In other words, persister cells may not respond to fluoroquinolones or express SulA due to their dormant state. However, two independent groups analyzed the SOS response in ofloxacin-treated Escherichia coli cultures at single-cell resolution [bib_ref] Single-cell imaging and characterization of Escherichia coli persister cells to ofloxacin in..., Goormaghtigh [/bib_ref] and found that antibiotic-induced DNA damage is similar in both persisters and antibiotic-sensitive cells. These findings demonstrate that persister cells can respond to external factors and indicate the presence of unique physiological activities in these cells that may be essential for their survival and resuscitation.
Studies of persisters are based on the premise that if a proposed mechanism is essential for the persister phenotype, genetically perturbing that mechanism should eliminate or reduce persister abundance. Such cells are quantified by persistence assays (e.g., clonogenic survival assays) in which culture samples are collected at various intervals during antibiotic treatment, washed, and plated on standard growth medium to enumerate surviving cells that can colonize in the absence of antibiotics [bib_ref] Definitions and guidelines for research on antibiotic persistence, Balaban [/bib_ref]. Unfortunately, this standard method does not distinguish persister formation mechanisms from resuscitation mechanisms. Thus, it is unclear from previous studies whether targeting Lon protease chemically or genetically eradicates persister cells or simply converts them to a viable but nonculturable (VBNC) state. To address this question, in the current study, we used vector constructs that allow fine-tuning of recombinant protein expression to verify that lon deletion (Dlon) impairs the resuscitation of ofloxacin (OFX) persisters by inhibiting FtsZ-dependent ring formation. We further showed that the reduction of persisters among lon-deficient cells can be transient based on environmental conditions, such as starvation. Indeed, starvation-induced SulA degradation or expression of Lon during the recovery period (i.e., after removal of antibiotics) restores the ability of nonculturable Dlon cells to form colonies by facilitating Zring formation, which represents a potential biomarker for Dlon persister cells transitioning to the normal cell state.
# Results
Lon is required for resuscitation of fluoroquinolone persisters. Fluoroquinolone antibiotics such as OFX inhibit DNA gyrase, leading to formation of double-stranded DNA breaks and induction of DNA repair mechanisms in persister cells [bib_ref] Single-cell imaging and characterization of Escherichia coli persister cells to ofloxacin in..., Goormaghtigh [/bib_ref] [bib_ref] Induction of the SOS response by new 4-quinolones, Phillips [/bib_ref]. Consistent with previous studies [bib_ref] Tolerance of Escherichia coli to fluoroquinolone antibiotics depends on specific components of..., Theodore [/bib_ref] [bib_ref] Highly contingent phenotypes of lon protease deficiency in Escherichia coli upon antibiotic..., Matange [/bib_ref] , we found that Dlon in E. coli MG1655 cells also significantly reduces levels of persisters compared to those found in a wild-type (WT) strain in cell cultures treated with OFX for 6 h [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref] in the supplemental material). While deletion of sulA from the WT strain (DsulA) did not affect OFX persister levels in E. coli, its deletion from the Dlon strain (DlonDsulA) restored persister levels of the lon-deficient cells [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref] , further verifying the reproducibility of previous studies [bib_ref] Tolerance of Escherichia coli to fluoroquinolone antibiotics depends on specific components of..., Theodore [/bib_ref] [bib_ref] Highly contingent phenotypes of lon protease deficiency in Escherichia coli upon antibiotic..., Matange [/bib_ref]. Using a sulA reporter (pMSs201-P sulA -gfp) in which the sulA promoter (P sulA ) is fused to a gene encoding green fluorescent protein (GFP) [bib_ref] A comprehensive library of fluorescent transcriptional reporters for Escherichia coli, Zaslaver [/bib_ref] , we assessed sulA expression in both WT and Dlon cells after OFX treatment [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref]. Although we identified a subset of OFX-treated cells that could not express GFP, we detected a significant number of GFP-positive cells expressing sulA in both the WT and Dlon strains, containing the sulA-gfp reporter plasmid (pMSs201-P sulA -gfp) were grown to stationary phase and diluted 100-fold in fresh LB medium. Diluted cells with increased filamentation observed in lon-deficient cells, an expected morphological feature mediated by SulA accumulation [bib_ref] SulA-dependent hypersensitivity to high pressure and hyperfilamentation after high-pressure treatment of Escherichia..., Aertsen [/bib_ref] [bib_ref] Role of sulA and sulB in filamentation by lon mutants of Escherichia..., Gottesman [/bib_ref] [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref].
To determine whether increased accumulation of SulA converts Dlon persisters to VBNC cells, we overexpressed Lon protease from a low-copy-number plasmid in lon-deficient cells during the recovery period on agar plates. This plasmid expression system was constructed using a cassette containing the isopropyl b-D-1-thiogalactopyranoside (IPTG)-inducible T5 promoter, lon, and the strong LacI q repressor, which was integrated into a pUA66 plasmid variant, thereby generating pUA66-lon. An empty vector (EV) without lon and a noninduced condition with pUA66-lon served as controls. Stationary-phase Dlon cells harboring either pUA66-lon or EV were diluted in fresh medium and treated with OFX for 6 h. Lon expression was not induced before or during treatment. OFX-treated cells were then collected, washed with sterile phosphate-buffered saline (PBS) solution to remove the antibiotic, and spotted onto Luria-Bertani (LB) agar plates containing different concentrations of IPTG to induce lon expression. We found that Lon overexpression during recovery rescues growth of OFX persisters in a concentration-dependent manner; this increase in colony-forming ability was not observed in the absence of IPTG or with the EV control [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref]. The same phenomenon was also observed in cell populations obtained from exponential-phase cultures [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref] to E). We verified that presence of the plasmid-based expression systems does not significantly alter the levels of persister cells observed in wild-type (WT) and Dlon strains . In addition, overexpressing Lon in OFX-treated WT cells during recovery did not significantly affect WT persistence . Next, to determine the role of Lon in E. coli persistence before and/or during OFX treatment, we added IPTG to cultures before and/or during treatment. Treated cells were then transferred to agar plates with or without IPTG for recovery [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref]. Although adding IPTG, and, thus, induction of Lon, at different stages did not impact WT persistence, Dlon persisters could only be rescued when IPTG was added in LB agar plates during the recovery period [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref]. These data further highlight the importance of Lon in OFX persister resuscitation.
SulA induction in the absence of OFX reduces culturability of the Dlon strain. Given that exposure to UV light induces both DNA damage and expression of SulA in both WT and Dlon strains [bib_ref] An inducible DNA replication-cell division coupling mechanism in E. coli, Huisman [/bib_ref] [bib_ref] Second site mutations in capR (lon) strains of Escherichia coli K-12 that..., Gayda [/bib_ref] , we next tested whether the fluoroquinolone-mediated phenomenon described above also occurs in cells exposed to UV in the absence of OFX. To this end, we diluted stationary-phase cells in fresh medium, exposed these diluted cells to UV light for various time intervals, and then plated the cells on solid medium to determine total CFU. Cells harboring the P sulA -gfp reporter were also subjected to UV exposure, cultured in liquid medium for 2 h, and examined microscopically to assess SulA expression. We found that similar to OFX treatment, UV exposure induced sulA expression in both WT and Dlon strains and increased filamentation of Dlon cells . Further, whereas our chosen time intervals of UV exposure did not affect WT cell viability or CFU levels , in response to this treatment, cells lacking Lon displayed significantly lower CFU than WT cells in a UV exposure time-dependent manner . Next, we tested whether Lon overexpression could rescue this colony formation deficiency. Following UV exposure, Dlon cells containing pUA66-lon or EV were plated on solid agar medium containing different concentrations of IPTG. As expected, we found that Lon expression eliminated the observed reduction in the culturability of Dlon cells in response to UV . We further detected a positive correlation between IPTG (lon inducer) and CFU levels to E), suggesting that recovery is dependent on Lon protein concentration, as well as on levels of accumulated SulA. were then treated with 5 mg/mL OFX for 6 h. After treatment, phase-contrast and fluorescence (GFP) micrographs of cells were obtained. Images at 0 (t = 0) and 6 h (t = 6 h) from a representative replicate are shown, but similar data were obtained from four biological replicates (n = 4). Scale bar, 25 mm. (C and D) E. coli cells (WT and Dlon) containing an inducible lon overexpression plasmid (pUA66-lon) or empty vector control (pUA66-EV) were treated with OFX (5 mg/mL) for 6 h and transferred to LB agar plates supplemented with isopropyl b-D-1-thiogalactopyranoside (IPTG) at the indicated concentrations. CFU were measured for each strain before and after treatment. n = 4. (E) WT and Dlon E. coli containing pUA66-lon were cultured with 1 mM IPTG before and/or during OFX treatment. After treatment, cells were transferred to agar plates with or without 1 mM IPTG. CFU were measured for each strain before and after treatment. n = 4. Statistical significance for pairwise comparisons was assessed using oneway analysis of variance (ANOVA) with Dunnett's post hoc test. **, P , 0.01; ***, P , 0.001; ****, P , 0.0001; ns, nonsignificant.
Both chemical (OFX) and environmental (UV) induction of SulA attenuated the colonyforming ability of lon-deficient cells [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref] , suggesting this phenomenon should be reproduced by plasmid-mediated SulA overexpression in the Dlon strain. To accomplish this, we generated a plasmid (pBAD-sulA) expressing sulA under the arabinose-inducible P BAD promoter and transferred this plasmid into both WT and Dlon strains harboring pUA66-lon or EV. We first noticed that leaky expression from pBAD-sulA inhibited growth of Dlon cells in liquid pretreatment cultures [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. Therefore, we supplemented both WT and Dlon pretreatment cultures with 0.25 mM IPTG to maintain Dlon cell growth. Both WT and Dlon strains harboring the plasmids were grown to stationary phase, washed to remove inducer, and then plated on solid medium containing arabinose (sulA inducer) and/or IPTG (lon inducer) at various concentrations to differentially express SulA and/or Lon during colony formation. Notably, although we did not fine-tune expression levels of these proteins on plates, we observed reduced CFU of WT cells and almost no CFU of the Dlon strain (under the limit of detection) at higher arabinose concentrations (.10 mM) [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. Further, whereas lower arabinose concentrations (,5 mM) did not affect the colony-forming ability of WT cells, colony formation of the Dlon strain was significantly compromised [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref] but reversed upon addition of IPTG [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref] , similar to what we observed for cells exposed to OFX and UV [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref].
Starvation rescues Dlon persisters. Persister cells are typically quantified using clonogenic survival assays in which antibiotic-treated cells are plated on agar medium and then incubated for at least 16 h. Therefore, we next tested whether a longer incubation period is Lon overexpression rescues the colony-forming ability of Dlon cells exposed to UV. (A) WT and Dlon E. coli containing pMSs201-P sulA -gfp were grown to stationary phase, diluted 100-fold in fresh LB medium, and exposed to UV for the indicated times. After exposure, cells were cultured in a shaker for 2 h, and phase-contrast and fluorescence (GFP) micrographs of cells were obtained. Images from a representative replicate are shown, but similar data were obtained for four biological replicates. n = 4. Scale bar, 25 mm. (B) WT and Dlon E. coli were exposed to UV as described in panel A and spotted onto LB agar plates to enumerate CFU. n = 4. (C to E) Cells containing the inducible lon overexpression plasmid (pUA66-lon) or empty vector (pUA66-EV) control were exposed to UV for the indicated times and transferred to agar plates supplemented with IPTG at indicated concentrations to enumerate CFU. n = 4. Statistical significance for pairwise comparisons was assessed using one-way ANOVA with Dunnett's post hoc test. ***, P , 0.001; ****, P , 0.0001; ns, nonsignificant.
Persistence State Mediated by Lon Protease ® necessary for Dlon colony formation [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. We found that, although a small number of both WT and Dlon colonies emerged at later time points and that existing colonies grew larger with longer incubation times, the 2-to 3-log difference in CFU between the WT and Dlon strains did not change with longer incubations [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. We made a similar observation for strains harboring pUA66-lon or EV that were plated on agar medium lacking IPTG coli with or without the inducible lon overexpression plasmid (pUA66-lon) or empty vector control (pUA66-EV) were grown to stationary phase and diluted 100-fold in fresh LB medium. Diluted cells were then treated with OFX for 6 h. After treatment, cells were washed, serially diluted, and spotted onto LB agar plates with or without IPTG (1 mM). Plates were incubated for 48 h, and images were captured at indicated time points. A representative replicate is shown, but similar data were obtained for four biological replicates. n = 4. (B) WT and Dlon E. coli with or without pUA66-lon or pUA66-EV were treated with OFX for 6 h, transferred to sterile PBS solution with or without 1 mM IPTG, and incubated at 37°C in a shaker for 7 days. Samples were collected at the indicated time points and spotted onto agar plates to enumerate surviving cells. n = 4. Statistical significance for pairwise comparison was assessed using one-way ANOVA with Dunnett's post hoc test. *, P , 0.05; **, P , 0.01; ***, P , 0.001; ns, nonsignificant. [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. Thus, while IPTG-induced Lon during recovery on agar medium could rescue Dlon cells harboring pUA66-lon [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref] , it was not clear whether Dlon cells that could not form colonies in the absence of IPTG [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref] were truly alive. To further investigate the ability of Dlon cells to form colonies and assess the resilience of Dlon persisters, cells with or without expression vectors were transferred to PBS solution after OFX treatment [bib_ref] Timing of DNA damage responses impacts persistence to fluoroquinolones, Mok [/bib_ref] [bib_ref] Viable but non-culturable and persistence describe the same bacterial stress state, Kim [/bib_ref] , and their colony-forming ability was tested daily by plating samples on agar medium with or without IPTG for 7 days [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. We found that persister subpopulations from both WT and Dlon strains were alive and able to survive for at least the 7-day duration of the experiment. Surprisingly, Dlon cells, even those without any expression vectors, could be gradually resuscitated when incubated in PBS solution [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref].
Given that SulA can be targeted by proteases other than Lon [bib_ref] ATP-dependent degradation of SulA, a cell division inhibitor, by the HslVU protease..., Seong [/bib_ref] and that starvation may further enhance its intracellular degradation [bib_ref] Viable but non-culturable and persistence describe the same bacterial stress state, Kim [/bib_ref] [bib_ref] Role of protein degradation in the survival of carbon-starved Escherichia coli and..., Reeve [/bib_ref] [bib_ref] Protein degradation in Escherichia coli. II. Strain differences in the degradation of..., Nath [/bib_ref] , we measured SulA concentrations in OFX-treated Dlon cell populations (including dead, VBNC, and persister cells) at days t = 0 and t = 2 during incubation in PBS solution and confirmed a decrease in SulA levels over time. Because persister cells were scarce, we used high-cell-density cultures for persister cell enrichment, which were generated using a procedure (see Materials and Methods) that does not eliminate the observed phenotypic switch during starvation but increases the number of resuscitated persister cells . To test if protein degradation is a general characteristic of starved cells, we used WT and Dlon cells harboring pMSs201-P sulAgfp and monitored GFP levels in OFX-treated cells during starvation. Although the antibiotic induced GFP expression from the sulA promoter, cellular GFP started to decrease when the OFX-treated cells were starved in PBS solution . Notably, we identified a subset of cells that did not respond to OFX or express GFP, consistent with our data in [fig_ref] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells [/fig_ref] and B. These cells might have been dead residual cells and/or VBNC cells; their condition remains to be determined. To show that the observed GFP reduction is not solely attributed to leakage of the proteins through damaged membranes, we stained the cells with propidium iodide (PI), Starvation enhances cellular protein degradation. (A and B) Stationary-phase cells of the Dlon strain were diluted 10-fold in fresh LB medium and treated with OFX for 6 h. Treated cells were collected, washed, and transferred to sterile PBS solution and cultured at 37°C in shaker for 2 days. Western blotting was performed at indicated time points (t = 0 and 2 days) to measure SulA expression (19 kDa). CFU counts were obtained before and after PBS incubation. n = 3. (C) The WT and Dlon strains with pMSs201-P sulA -gfp were transferred to PBS solution after OFX treatment and cultured at 37°C with shaking for 7 days. Cells were analyzed with a flow cytometer for GFP measurements at indicated time points. A representative replicate is shown, but similar data were obtained for four biological replicates. n = 4. (D) OFX-treated cells described in panel C were stained with propidium iodide (PI) and analyzed with a flow cytometer. Live cells and ethanol (70% [vol/vol])-treated cells (dead cells) served as negative and positive controls, respectively. A representative replicate is shown, but similar data were obtained for four biological replicates. n = 4. Pairwise statistical significance was performed using two-tailed t test with unequal variance, where **, P , 0.01. Persistence State Mediated by Lon Protease ® which does not permeate cells with intact membranes. PI staining verified that a significant number of intact WT and Dlon cells still exhibited increased GFP degradation after starvation . Also, we observed an enrichment of intact cells at later time points of PBS incubation, as some dead cells were eventually lysed during PBS incubation . Altogether, our results imply that the resuscitation of Dlon cells under starvation conditions might result from increased SulA degradation.
PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures. Given that Dlon potentially converts persisters to VBNC cells, we employed flow cytometry and an IPTG-inducible gfp expression system (pUA66-gfp) to monitor persister cell resuscitation. After OFX treatment, the WT and Dlon strains harboring pUA66-gfp were transferred to fresh LB recovery medium (liquid) containing IPTG. We note that GFP was not induced in pretreatment cultures or during OFX treatment. When persister cells are resuscitated and then proliferate in recovery cultures, the cells should express GFP in the presence of IPTG and be detectable with flow cytometry [bib_ref] Flow-cytometry analysis reveals persister resuscitation characteristics, Mohiuddin [/bib_ref]. Interestingly, our analysis revealed that a large number of OFX-treated cells in both WT and Dlon recovery cultures started to express GFP [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] , subpopulations highlighted with green circles) while preserving their membrane integrity [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref]. Although the GFP-expressing WT cells in recovery cultures corresponded to ;3.28 6 0.53% of the initial cell population (before OFX treatment), the percentage of these cells was much higher in the Dlon recovery cultures (;7.31 6 1.46%). Given that these cells can express GFP and that only a small fraction of them (i.e., persisters) could exit the nongrowth state and proliferate upon their transition to fresh medium (WT persisters, 0.125 6 0.032% of the initial population; Dlon persisters, ;0.0015 6 0.001% of the initial population), we presume that these GFP-positive cells largely exhibit VBNC phenotypes. The proliferating subpopulation in the WT recovery culture became more noticeable upon flow cytometry at approximately 7 to 8 h after transfer to new medium [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] , subpopulation highlighted with a red circle), whereas this subpopulation was not detected in the Dlon recovery culture throughout the course of the study [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref]. However, when starved in PBS solution for 2 days, Dlon persisters could be resuscitated in recovery cultures similar to WT cells [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] , subpopulations highlighted with a red circle). Although we did not observe many GFP-expressing cells after starvation, the recovery cultures still contained a significant number of intact cells [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] whose size started to increase upon their transfer to fresh medium, which was verified by their increased forward scatter (FSC-H) signals [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] , subpopulations highlighted with orange circles). Although it is not clear whether these intact cells are truly VBNC cells (which is beyond the study's scope), our data further showed that incubation in PBS solution facilitated persister resuscitation in both WT and Dlon recovery cultures. Specifically, starved persister cells were resuscitated in less time (2 to 3 h) [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] than persisters from unstarved cell cultures [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref]. In addition, the resuscitating cells seemed to be elongated after starvation, and their size (i.e., FSC-H) considerably increased [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] , subpopulations highlighted with red circles). Altogether, these results verify the existence of a phenotypic switch between a nonculturable to a culturable cell state in OFX-treated Dlon cells that can be facilitated by starvation conditions.
Starvation facilitates Z-ring formation in OFX-treated Dlon cells. Although SulA accumulation in lon-deficient cells is expected to inhibit Z-ring formation [bib_ref] Inhibition of FtsZ polymerization by SulA, an inhibitor of septation in Escherichia..., Mukherjee [/bib_ref] , this process has not been investigated in Dlon persisters. Z-ring formation at the possible division site of a bacterium takes place when FtsZ, a filamentous tubulin-like protein, assembles into a ring shape [bib_ref] Bacterial cell division and the Z ring, Lutkenhaus [/bib_ref] [bib_ref] Z-ring force and cell shape during division in rod-like bacteria, Lan [/bib_ref] [bib_ref] A widely conserved bacterial cell division protein that promotes assembly of the..., Gueiros-Filho [/bib_ref]. To study this event in Dlon persisters, we generated and validated a lowcopy expression system (pUA66-ftsZ-gfp) in which ftsZ is fused with gfp (46) and controlled by the IPTG-inducible T5 promoter [bib_ref] A multi-layered protein network stabilizes the Escherichia coli FtsZ-ring and modulates constriction..., Buss [/bib_ref]. Here, using this FtsZ-GFP construct, we observed Zring formation in exponentially growing WT cells of various sizes and shapes, such as smaller cells with single rings and filamentous cells with randomly or orderly spaced multiple rings (data available upon request), as reported elsewhere [bib_ref] A multi-layered protein network stabilizes the Escherichia coli FtsZ-ring and modulates constriction..., Buss [/bib_ref] [bib_ref] Regrowth-delay body as a bacterial subcellular structure marking multidrug-tolerant persisters, Yu [/bib_ref] [bib_ref] Identification and characterization of a negative regulator of FtsZ ring formation in..., Levin [/bib_ref]. We also confirmed that FtsZ-GFP expression did not affect the observed 2-to 3-log difference between WT and Dlon persistence (data available upon request).
To investigate cellular Z-ring formation in OFX-treated cultures, cells harboring the FtsZ reporter were transferred to an LB agarose pad after OFX treatment and then monitored with fluorescence microscopy. OFX-treated WT cells showed highly heterogeneous cell sizes and morphologies, (e.g., smooth and rough cells) [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. FtsZ-GFP proteins were primarily aggregated in rough cells [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref] , which were potentially dead, as their membranes were highly damaged . FtsZ assemblies were generally found to be structurally heterogeneous in smooth, intact cells and included regularly spaced multiple Z-rings or linear, spiral, elliptical, and "8"-shaped structures that may have been transitioning to a ring shape [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. Similar to the WT strain, the heterogeneous FtsZ assemblies and cell morphologies were Persistence State Mediated by Lon Protease ® also observed in the DsulA and DlonDsulA strains [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref]. Conversely, Z-ring formation was rarely seen in Dlon cell populations, which were instead enriched with cells in which FtsZ was dispersed in smooth cells or aggregated in rough cells [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. However, when starved in PBS solution for 2 days, OFX-treated Dlon cultures formed cell subpopulations with structurally heterogeneous FtsZ assemblies, including randomly or orderly spaced multiple rings and linear or spiral filamentous structures [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. When we monitored OFX-treated WT, Dlon, DsulA, and DlonDsulA cells on LB agar pads with timelapse microscopy after starvation, we noted that healthy, elongated cells of all strains containing multiple highly organized Z-rings could be resuscitated within 2 to 4 h [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. These cells first exhibited an extensive elongation, consistent with our flow cytometry data [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref] , and then divided at the septal points (49) [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. Notably, we did not observe were grown to stationary phase and diluted 100-fold in fresh LB medium. Diluted cells were then treated with OFX for 6 h. IPTG (1 mM) was added to cultures before and during treatment to express FtsZ-GFP. After treatment, cells were collected, washed with PBS solution to remove the antibiotic, and spread on an LB agarose (1%) pad. The pad was then monitored with a fluorescence microscope to collect phasecontrast and fluorescence images. The magnified FtsZ assemblies highlighted by numbers are provided in the right panel. (C) Dlon cells harboring pUA66-ftsZ-gfp were transferred to PBS solution after OFX treatment and cultured for 2 days at 37°C with shaking (250 rpm). Then, cells were collected and transferred to pads to monitor FtsZ assemblies. Images from a representative replicate are shown, but similar data were obtained for four biological replicates. n = 4. Scale bar, 25 mm.
persister cells in which FtsZ polymerization underwent a structural change from linear or spiral to ring-shaped assemblies [bib_ref] Z-ring force and cell shape during division in rod-like bacteria, Lan [/bib_ref] [bib_ref] Identification and characterization of a negative regulator of FtsZ ring formation in..., Levin [/bib_ref]. While shorter cells with fewer rings were rarely resuscitated, cells with aggregated proteins or dispersed GFP could not be resuscitated at all. Of note, protein aggregation was not due to the overexpression of FtsZ from the low-copynumber plasmids, we still observed aggregation phenotypes in OFX-treated cells without any expression vector in phase-contrast micrographs (data available upon request). Also, detection of resuscitating persisters in OFX-treated Dlon cells with microscopy was extremely difficult when the cells were not starved in PBS.
Z-ring architecture is a key biomarker of Dlon persisters. Although starvation facilitated persister resuscitation and Z-ring formation in the Dlon strain, persister cells still represented only a small fraction of the intact cell subpopulation in antibiotic-treated cultures. To determine whether the Z-ring is a suitable biomarker for Dlon persisters during their transition to a normal cell state, we monitored hundreds of intact but diverse (in cell shape, size, and Zring architecture) OFX-treated Dlon cells with time-lapse microscopy after PBS starvation. Cell lengths, as well as Z-ring structures and numbers, were determined using phase-contrast and fluorescence microscopy directly after cells were transferred to microscope pads. Resuscitated cells were identified using time-lapse microscopy. Our in-depth image analysis revealed that persister resuscitation in the Dlon strain strongly correlated with cell size and the number of Zrings [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref]. Specifically, cells with linear or spiral FtsZ assemblies could not be resuscitated [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref]. Moreover, when we calculated the ratio of cell length in microns (L) to the number of Z-rings (Z) for each cell analyzed, we found a correlation between persister resuscitation and calculated L/Z values [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref] , which was supported by binomial regression analysis [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref]. Cells with L/Z % 1, L . 5 mm, and Z . 5 were generally resuscitated [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref]. While L/Z ratios of nonresuscitated cells were often much greater than 1 [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref] , we observed some nonresuscitating phenotypes in cells with L/Z % 1. However, these cells were generally smaller (L , 5 mm) and had fewer Z-rings (Z , 5) than persister cells [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref]. Although we did not perform this labor-intensive analysis for the DsulA and DlonDsulA strains, we were able to report similar results for the WT strain [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref] , verifying the existence of a conserved relationship between persister resuscitation and Z-ring architecture in E. coli MG1655 under the conditions studied here.
The capability of persister cells to form highly organized multiple Z-rings [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref] and [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref] should be enhanced by starvation [fig_ref] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures [/fig_ref]. We have also noticed that sulA deletion induces thick and well-defined Z-ring structures in persister cells [fig_ref] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells [/fig_ref]. As the sulA promoter (P sulA ) is tightly regulated by a RecA-LexA-dependent global DNA damage response mechanism, we presume that persisters (unlike VBNC cells) can successfully repair OFX-induced DNA damage, thus reducing SulA expression during the recovery period; this mechanism may explain the observed Z-ring structures in these cells [fig_ref] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker [/fig_ref]. Although the DNA repair mechanism(s) in persisters was not the main focus of this study, we analyzed P sulA activity of OFXtreated WT cells (harboring pMSs201-P sulA -gfp) in recovery cultures with fluorescence microscopy and flow cytometry. While most resuscitating WT cells were initially SulA positive after OFX treatment, their sulA promoter activity decreased when they began to elongate and divide in fresh, antibiotic-free medium . We also identified a cell subpopulation that continued to express SulA while elongating; however, these cells did not divide throughout the course of the study , implying the presence of DNA damage. Our flow cytometry analysis also showed that a large number of resuscitating cells had reduced sulA promoter activity . Although we did not monitor P sulA activity of OFX-treated cells after PBS starvation, as the GFP variant from pMSs201-P sulA -gfp was mostly degraded , the lack of DNA damage response in daughter cells highlights the ability of persister cells to efficiently repair OFX-induced cellular damage, which is consistent with a previously published study.
Lon overexpression during recovery facilitates Z-ring formation in OFX-treated Dlon cells. Finally, to demonstrate if Lon overexpression during recovery (without starvation) facilitates Z-ring formation in lon-deficient cells, we generated a pBAD-ftsZ-gfp plasmid (expressing the FtsZ reporter under the control of an arabinose-inducible promoter) and introduced it to the Dlon strain harboring pUA66-lon. Because growth of the Dlon strain with both expression vectors was markedly decreased on microscope pads, we performed resuscitation experiments in liquid cultures. Lon expression was not induced in cultures before and during antibiotic treatment. After OFX exposure, cells were transferred to fresh liquid medium with or without IPTG (the lon inducer), and at designated time points, samples from the resuscitation cultures were collected and plated on solid medium (with or without IPTG). We found that Dlon cells started to resuscitate approximately 8 h after their transfer to fresh liquid medium with IPTG, as evidenced by increased CFU levels due to cell proliferation near the same time point . As expected, the resuscitation of Dlon persisters was impaired in the absence of IPTG , and similar CFU profiles were obtained for Dlon cells containing the vector control . We also investigated the collected samples with fluorescence microscopy and found that, despite their scarcity, persister cells began to form Z-rings after 8 h of culturing in the presence of IPTG and exhibited highly heterogeneous morphologies . FtsZ assemblies were also found to be structurally heterogeneous (e.g., linear and spiral structures) . Conversely, this phenomenon was rarely observed in Dlon cultures in the absence of IPTG; rather, these cultures were enriched with nonresuscitating cells in which FtsZ was typically dispersed or aggregated .
# Discussion
The Lon protease has been shown to degrade misfolded proteins, RNases, heat shock proteins, and transfer-messenger RNA (tmRNA)-associated proteins, as well as components of chromosomal and/or plasmid-based toxin/antitoxin systems [bib_ref] SulA-dependent hypersensitivity to high pressure and hyperfilamentation after high-pressure treatment of Escherichia..., Aertsen [/bib_ref] [bib_ref] Functional dissection of a cell-division inhibitor, SulA, of Escherichia coli and its..., Higashitani [/bib_ref] [bib_ref] Regulation of cell division in Escherichia coli: SOS induction and cellular location..., Schoemaker [/bib_ref] [bib_ref] Capsule synthesis in Escherichia coli K-12 is regulated by proteolysis, Torres-Cabassa [/bib_ref] [bib_ref] Structural inhibition and reactivation of Escherichia coli septation by elements of the..., Dopazo [/bib_ref] [bib_ref] The product of the lon (capR) gene in Escherichia coli is the..., Chung [/bib_ref] [bib_ref] Degradation by proteases Lon, Clp and HtrA, of Escherichia coli proteins aggregated..., Laskowska [/bib_ref]. Although these known functions and previously published data suggest Lon is involved in persister formation, our results have identified a crucial role for Lon in the resuscitation and recovery of persister cells. The work in this area, including the current study, suggests that SulA-Lon acts as a toxin-antitoxin module during quinolone treatment (SulA as a toxin) [bib_ref] Tolerance of Escherichia coli to fluoroquinolone antibiotics depends on specific components of..., Theodore [/bib_ref] [bib_ref] Highly contingent phenotypes of lon protease deficiency in Escherichia coli upon antibiotic..., Matange [/bib_ref]. However, sulA only appears to be relevant in the absence of lon, as both the DsulA and double mutant DlonDsulA strains showed persister levels similar to those of WT cells. Nevertheless, Lon may still be an attractive target for small molecular inhibitors [bib_ref] Leveraging peptide substrate libraries to design inhibitors of bacterial lon protease, Babin [/bib_ref] , as it is a critical protein for bacterial cell survival.
One way to differentiate persister formation mechanisms from those involved in persister resuscitation is to controllably express the genes of interest in their respective mutant strains before, during, and after antibiotic treatment. If a gene is essential for persister resuscitation, its expression in the mutant strain during recovery should be sufficient to restore the colony-forming ability of antibiotic-treated cells, as shown in this study, in which the inducible expression of Lon in Dlon cells during recovery from OFX treatment increased their culturability. However, culturability of Dlon persisters can also be restored when the cells are starved after OFX treatment. Interestingly, this phenomenon was also observed in WT cells, as we detected an increase (;10-fold) in persisters of WT cultures during starvation [fig_ref] FIG 3: Starvation in PBS solution rescues Dlon persisters [/fig_ref]. These results are consistent with those of previous studies [bib_ref] Timing of DNA damage responses impacts persistence to fluoroquinolones, Mok [/bib_ref] [bib_ref] Toxin induction or inhibition of transcription or translation posttreatment increases persistence to..., Lemma [/bib_ref] , which found that starvation, toxin induction, and chemical inhibition of transcription or translation following antibiotic treatment rescued E. coli persisters. Although these stressors delay growth-related processes which are thought to facilitate DNA repair and survival [bib_ref] Timing of DNA damage responses impacts persistence to fluoroquinolones, Mok [/bib_ref] [bib_ref] Toxin induction or inhibition of transcription or translation posttreatment increases persistence to..., Lemma [/bib_ref] , our data suggest that SulA degradation might also play a crucial role in persister rescue. However, these two proposed mechanisms are not mutually exclusive; in fact, they may occur in the cells simultaneously, as both are necessary for persister cell resuscitation.
We measured SulA concentrations in OFX-treated Dlon cells, which included the entire cell populations of dead, VBNC, and persister cells. Unfortunately, direct measurement of persister cell physiology is challenging due to the difficulty in isolating pure samples. Many technical challenges arise from their low abundance, transient nature and similarities to VBNC cells, which are more abundant than persister cells. Given that (i) starvation enhances intracellular protein degradation [bib_ref] Viable but non-culturable and persistence describe the same bacterial stress state, Kim [/bib_ref] [bib_ref] Role of protein degradation in the survival of carbon-starved Escherichia coli and..., Reeve [/bib_ref] [bib_ref] Protein degradation in Escherichia coli. II. Strain differences in the degradation of..., Nath [/bib_ref] , (ii) SulA is a fairly unstable protein [bib_ref] ATP-dependent degradation of SulA, a cell division inhibitor, by the HslVU protease..., Seong [/bib_ref] , and (iii) cellular GFP was completely degraded in the entire cell population during incubation in PBS , the resuscitation of Dlon cells under starvation conditions potentially results from SulA degradation. This finding is further supported by our aforementioned experiments that show chemical (OFX) and environmental (UV) induction of SulA and plasmid-mediated SulA overexpression attenuate the colony-forming ability of lon-deficient cells.
One key insight from this study relates to issues associated with VBNC cells, which are generally more abundant than persister cells, can be metabolically active, stain as living cells, and preserve their membrane integrity [bib_ref] Establishment of a method to rapidly assay bacterial persister metabolism, Orman [/bib_ref]. Further, although they also survive antibiotic treatments, their resuscitation in standard medium is typically not observed [bib_ref] Bridging the gap between viable but non-culturable and antibiotic persistent bacteria, Ayrapetyan [/bib_ref]. In general, the clinical importance of VBNC cells of pathogenic bacteria is recognized, as certain bacteria (e.g., Mycobacterium tuberculosis) are known to survive in the human body for years without growing or causing symptoms [bib_ref] Mechanisms of latency in Mycobacterium tuberculosis, Parrish [/bib_ref] [bib_ref] Persistent bacterial infections, antibiotic tolerance, and the oxidative stress response, Grant [/bib_ref]. These cells are fastidious under laboratory conditions and are most effectively detected and identified by PCR and immunological methods [bib_ref] Persistent bacterial infections, antibiotic tolerance, and the oxidative stress response, Grant [/bib_ref] [bib_ref] Immunologic diagnosis of tuberculosis: a review, Chan [/bib_ref]. Critically, although perturbation of certain mechanisms may reduce persister levels in a cell population, VBNC cells may not be eliminated, which poses an important health concern. In this study, our data show that environmental signals can trigger VBNC cell resuscitation. Therefore, measuring VBNC cell levels in cultures is essential to better understand the mechanism by which they can resume growth. We previously developed flow cytometry techniques for this purpose [bib_ref] Establishment of a method to rapidly assay bacterial persister metabolism, Orman [/bib_ref] [bib_ref] Identifying metabolic inhibitors to reduce bacterial persistence, Mohiuddin [/bib_ref] [bib_ref] Dormancy is not necessary or sufficient for bacterial persistence, Orman [/bib_ref] and showed here that resuscitation of VBNC Dlon cells occurs under starvation conditions.
Another important finding from our study helps shed light on the long-standing unsolved question of why only a small fraction of intact cells can be resuscitated after the removal of antibiotics. Although microscopy, which has been used extensively by many research groups to study rare persister cells [bib_ref] Single-cell imaging and characterization of Escherichia coli persister cells to ofloxacin in..., Goormaghtigh [/bib_ref] [bib_ref] Enrichment of persisters enabled by a ß-lactaminduced filamentation method reveals their stochastic..., Windels [/bib_ref] [bib_ref] Bacterial persistence as a phenotypic switch, Balaban [/bib_ref] , can provide single-cell resolution, this approach cannot be reliably applied to distinguish persisters from VBNC cells without a persister-specific biomarker. This limitation is because both types of cells are in a growth-inhibited state during antibiotic treatment and can only be discriminated after persisters exit their persistence state and start to replicate following antibiotic removal. Resuscitation mechanisms may be regulated in a stochastic and threshold manner in persister cells during their transition to a normal cell state [bib_ref] Enrichment of persisters enabled by a ß-lactaminduced filamentation method reveals their stochastic..., Windels [/bib_ref] , but these mechanisms can only be elucidated if persister cells are detected and characterized during their transition and before they revert to normal, proliferating cells. Notably, our findings here suggest that Z-ring formation may be a key biomarker for distinguishing these cells.
Overall, the results of this study show that Lon plays a critical role in persister cell resuscitation through a mechanism dependent on SulA and FtsZ. We further demonstrate that the nonculturable state of antibiotic-tolerant Dlon cells is transient, as their colony-forming ability can be restored by starvation, which likely leads to SulA degradation. Finally, we show that starvation-induced SulA degradation or expression of Lon during recovery facilitates Zring formation in Dlon persisters, suggesting altered Z-ring architecture may be a biomarker for persister cells transitioning to a normal cell state.
# Materials and methods
Bacterial strains and plasmids. All experiments were conducted using E. coli MG1655 and its derivative strains. E. coli MG1655 and the pQE-80L and pUA66 plasmids were obtained from Mark P. Brynildsen at Princeton University, and pXY027 was a gift from Jie Xiao (plasmid no. 98915; Addgene, Watertown, MA, USA) [bib_ref] A multi-layered protein network stabilizes the Escherichia coli FtsZ-ring and modulates constriction..., Buss [/bib_ref]. The Dlon, DsulA, and DlonDsulA strains were generated using the Datsenko-Wanner method, as described previously [bib_ref] One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products, Datsenko [/bib_ref]. The pQE-80L plasmid has an isopropyl b-D-1-thiogalactopyranoside (IPTG)-inducible synthetic T5 promoter and a strong constitutive LacI q repressor. The pUA66-gfp plasmid variant was generated by transferring the T5-gfp-lacI q segment from pQE-80L-gfp into pUA66 that has kanamycin resistance gene (Kan r ). Removal of gfp from pUA66-gfp led to development of a modified pUA66 empty vector. The lon gene was then cloned into the modified pUA66 plasmid to generate the pUA66-lon expression system. The SulA reporter (pMSs201-P sulA -gfp) was generated in a previous study [bib_ref] A comprehensive library of fluorescent transcriptional reporters for Escherichia coli, Zaslaver [/bib_ref]. A pBAD plasmid variant containing the ampicillin resistance gene (Amp r ), an arabinose-inducible promoter (P BAD ), the araC gene encoding the P BAD inhibitor, and pBRR322ori were obtained from Thermo Fisher Scientific (catalog no. V44001; Waltham, MA, USA). The sulA gene was cloned into this pBAD plasmid to generate the pBAD-sulA expression system. To generate the pBAD-ftsZ-gfp expression system, a cassette containing the ftsZ gene fused with gfp, and a chloramphenicol resistance (Cm r ) gene was amplified from pXY027 and ligated to the P BAD -araC-pBR322ori cassette obtained from the pBAD plasmid. To generate the pAU66-ftsZ-gfp plasmid, the ftsZ-gfp DNA fragment, amplified from pXY027 plasmid, was cloned into the modified pUA66 empty vector. FtsZ-GFP expression systems were transferred to WT, Dlon, DsulA, and DlonDsulA strains that also contain the wild type ftsZ gene. All plasmids were generated using standard cloning methods (https://www.neb.com/tools-and-resources/feature-articles/foundations-of -molecular-cloning-past-present-and-future?__cf_chl_jschl_tk__=SNruurTMsKPXPp0X7YTcODunmJAHSq9MW9W NWyaasjM-1639784771-0-gaNycGzNCL0). Genetic modifications were verified by PCR and gene sequencing (Genewiz, South Plainfield, NJ, USA). A complete list of strains, plasmids, and oligonucleotides used in this study is presented in in the supplemental material.
Media, chemicals, and culture conditions. Unless stated otherwise, all chemicals, antibiotics, and enzymes used in these experiments were purchased from Thermo Fisher Scientific, VWR International (Radnor, PA, USA), Sigma-Aldrich (St. Louis, MO, USA), or New England Biolabs (Ipswich, MA, USA). All liquid media were prepared with ultrapure deionized (DI) water, having an electrical resistivity of 18.2 MX.cm. Luria-Bertani (LB) liquid medium was prepared by dissolving 10 g tryptone, 10 g sodium chloride, and 5 g yeast extract in 1 L ultrapure DI water. LB agar medium was prepared by dissolving 40 g premixed LB agar powder in 1 L ultrapure DI water. Phosphate-buffered saline (PBS, 1Â) solution was used to remove chemicals and antibiotics from bacterial cell cultures; this was prepared by mixing 10Â sterile PBS solution with autoclaved ultrapure DI water. Persister assays were performed with 5 mg/mL ofloxacin (OFX). The MIC of OFX for E. coli MG1655 was determined previously [bib_ref] Identifying metabolic inhibitors to reduce bacterial persistence, Mohiuddin [/bib_ref]. To select for plasmid maintenance, kanamycin (25 mg/mL), chloramphenicol (25 mg/mL), and ampicillin (50 mg/mL) were added to both liquid and solid cultures. To induce expression of Lon, FtsZ, and SulA, medium was supplemented with the inducer (IPTG and L-arabinose, depending on the plasmid used; see in the supplemental material for details) at the indicated concentrations. All chemicals and antibiotics added to bacterial cultures were dissolved in ultrapure DI water and sterilized by passage through 0.2-mm membrane filters. Solid and liquid media were sterilized by autoclaving. Overnight precultures and main cultures were prepared in test tubes containing 2-mL LB broth, and these were grown at 37°C with shaking (250 rpm) for 24 h. Overnight precultures were inoculated from cell stocks frozen in 25% glycerol at 280°C. Main cultures grown for 24 h were considered to be in stationary phase.
Cell growth and persister assays. To prepare main cultures, cells from overnight precultures were diluted 1:1,000 in 2-mL LB broth in 14-mL test tubes and cultured in a shaker for 24 h. Growth was assessed by measuring optical density at 600-nm wavelength (OD 600 ) using a Varioskan Lux multimode microplate reader (Thermo Fisher Scientific). Stationary-phase cells from the main cultures (t = 24 h) were diluted 100-fold in 2-mL LB in 14-mL test tubes and treated with OFX (5 mg/mL) for 6 h. At the indicated time points, 1 mL of culture was removed from the test tubes, collected into microcentrifuge tubes, and centrifuged at 13,000 rpm to pellet the surviving cells. After centrifugation, 900 mL of supernatant was removed from the tubes, and 900 mL PBS solution was added. This washing procedure was repeated at least twice to reduce the antibiotic concentration to sub-MIC levels. After the final centrifugation, 900 mL of supernatant were removed; cell pellets were resuspended in the remaining 100 mL of PBS solution, and concentrated cell suspensions were transferred to the wells of a round-bottom 96-well plate. Cell suspensions were then serially diluted (10-fold) in PBS solution, and 10 mL of the diluted cell suspensions were spotted onto agar plates to enumerate persister levels. When required, cells were spotted onto agar plates supplemented with IPTG (0.1 to 1 mM), L-arabinose (1 to 20 mM), kanamycin (25 mg/mL), chloramphenicol (25 mg/mL), or ampicillin (50 mg/ mL) at the indicated concentrations. To increase the limit of detection, the remaining 80 mL of the concentrated cell suspensions were plated. This serial dilution and plating procedure were also performed for cell cultures before antibiotic treatment to quantify the total number of cells in these cultures. To quantify persister cells in exponential-phase cultures, cells from an overnight culture were diluted (100-fold) in 25 mL LB in a 250-mL baffled flask, cultured until they reached an OD 600 of ;0.2, and then treated with OFX. At the indicated time points during treatment, cells were collected, washed, and plated to quantify CFU as described above. Plates were incubated at 37°C for at least 16 h to count CFU and then incubated up to an additional 48 h to see if new colonies emerged. Of note, we refer to overnight and main cultures as "pretreatment cultures" and OFX-treated cultures as "treatment cultures" in the manuscript. Solid and liquid cultures to which OFX-treated cells were transferred after removal of antibiotics are referred to as "recovery cultures." When indicated, the Lon expression was induced in pre-and treatment cultures with 1 mM IPTG.
UV exposure assay. WT and Dlon cells from stationary-phase cultures were diluted 1:100 in 1-mL LB, transferred to petri dishes, and exposed to UV light (UVP ChemStudio, catalog no. 849-97-0928-02; Analytik Jena, Jena, Germany) for different lengths of time (i.e., 0, 1, 2, and 4 min). After UV exposure, cells were collected, serially diluted in PBS solution, and then spotted onto agar plates (61 mM IPTG). Agar plates were incubated for at least 16 h at 37°C to enumerate CFU. Cells expressing SulA reporters were similarly exposed to UV; these cells were then transferred to test tubes and cultured for 2 h in a shaker prior to imaging analysis (described below).
SulA overexpression. WT and Dlon cells harboring the pBAD-sulA and pAU66-lon plasmids from stationary-phase cultures were serially diluted (10-fold) in PBS solution, and 10 mL of the diluted cell suspensions were spotted onto agar plates containing L-arabinose (the SulA inducer) at various concentrations (0, 1, 5, 10, and 20 mM) with/without 1 mM IPTG (the Lon inducer). Of note, pretreatment cultures of both WT and Dlon strains were supplemented with 0.25 mM IPTG for Lon expression to maintain the growth of Dlon cells, as the leaky expression of SulA from the pBAD-sulA plasmid impairs the growth of Dlon cells in liquid pretreatment cultures.
Starvation in PBS solution. Stationary-phase WT and Dlon cells were diluted 1:100 in 14-mL test tubes containing 2 mL LB and treated with OFX for 6 h. Treated cultures (2 mL) were collected, washed with PBS solution to remove antibiotics, resuspended in 2-mL sterile PBS solution, and cultured at 37°C with shaking (250 rpm) for 7 days. At the indicated time points (0, 1, 2, 3, 5, and 7 days), 100-mL cell samples were collected, serially diluted, and spotted onto agar plates (containing either 1-mM IPTG or no IPTG), as described above. Agar plates were incubated at 37°C for at least 16 h to enumerate CFU.
Western blotting for SulA protein detection in high-cell-density cultures. To quantify intracellular SulA protein levels, we used high-cell-density cultures. Briefly, stationary-phase cells were diluted 10-fold in 25 mL fresh LB medium in 250-mL flasks, treated with OFX for 6 h, and then washed with PBS solution to remove the antibiotic and resuspended in 25 mL PBS solution in flasks and cultured at 37°C with shaking (250 rpm). Cultures (including persister, VBNC, and dead cells) were collected into 50-mL falcon tubes on day 0 and day 2 during PBS starvation and centrifuged at 4,700 rpm for 15 min to sediment the cells, which were then frozen with dry ice and sent to RayBiotech (Peachtree Corners, GA, USA) for the SulA quantitation. Cells were also plated to quantify CFU. Cell lysis, protein quantitation, and detection were performed by RayBiotech scientists. Samples containing the same total protein concentration were analyzed by an automatic western machine, in which proteins are separated by size and immobilized in a capillary system. The target protein (SulA) was then chemiluminescently quantified using a primary antibody (rabbit anti-E. coli SulA polyclonal antibody; catalog no. MBS7004012; MyBioSource, San Diego, CA, USA) and a horseradish peroxidase-conjugated secondary antibody.
Monitoring GFP degradation. Stationary-phase cells of E. coli MG1655 WT and Dlon strains harboring the pMSs201-P sulA -gfp plasmid were diluted 100-fold in 2-mL LB in 14-mL test tubes and treated with OFX (5 mg/mL) for 6 h. After the treatment, the cells were washed with PBS solution to remove the antibiotic, and resuspended in PBS solution and cultured at 37°C with shaking (250 rpm). Cells at indicated time points during the PBS incubation were analyzed with a flow cytometer (NovoCyte flow cytometer; NovoCyte 3000RYB; ACEA Biosciences Inc., San Diego, CA). Cells were excited at 488 nm, and green fluorescence was detected with a 530/ 30 nm emission filter. Live cells with or without GFP expression systems were used as controls.
PI staining. OFX-treated cells were collected, washed with PBS solution, and resuspended in 0.85% NaCl solution to achieve a desired cell density for flow cytometry analysis (;10 6 cells/mL). After adding PI (20 mM), cells were incubated in dark at 37°C for 15 min. The cells were then analyzed with a flow cytometer. Unstained live cells were used to gate the cell population on the flow cytometry diagram using forward and side scatter parameters. PI-stained dead cells obtained from ethanol (70%) treatment were used as a positive control. PI-stained live cells were used as a negative control. Cells were excited at 561-nm wavelength, and red fluorescence was detected with a 615/20-nm bandpass filter. When necessary, PI-stained cells were also analyzed with fluorescence microscopy (see "Microscope imaging" section for details).
Monitoring persister cell resuscitation with flow cytometry. OFX-treated WT and Dlon cells harboring the pUA66-gfp plasmid were washed with PBS solution to remove the antibiotic and then resuspended in fresh LB medium containing 1 mM IPTG and cultured at 37°C with shaking (250 rpm). However, IPTG was not added in overnight and treatment cultures. At designated time points, cells were collected from the recovery culture and analyzed with a flow cytometer. Cells were excited at 488 nm, and green fluorescence was detected with a 530/30-nm emission filter.
Microscope imaging. (i) Z-ring imaging. Stationary-phase E. coli MG1655 WT, Dlon, DsulA, and DlonDsulA cells harboring the pUA66-ftsZ-gfp plasmid were diluted 1:100-fold in 2-mL LB broth in 14-mL test tubes and then treated with OFX for 6 h. IPTG (1 mM) was added to overnight and main cultures to induce ftsZgfp expression. Treated cells were then collected, washed to remove the antibiotic, transferred to 2 mL PBS solution in 14-mL test tubes, and cultured at 37°C with shaking (250 rpm). After starving the cells in PBS solution for 2 days, 1-mL cell suspensions were collected from PBS cultures, pelleted by centrifugation, and resuspended in 1 mL LB media. The resuspended cells were then transferred to LB agar pads, which were dried next to a Bunsen burner flame for 30 min. The agar pads were prepared by dissolving agarose (1%) in LB liquid medium followed by microwave sterilization. After cooling the agarose medium, kanamycin (25 mg/mL) and IPTG (1 mM) were added for plasmid retention and ftsZ-gfp expression. The LB agarose medium was then poured over a glass slide (25 by 75 mm), each side of which was enclosed by stacked slides to make the pad smooth and sufficiently thick [bib_ref] Measuring mRNA copy number in individual Escherichia coli cells using single-molecule fluorescent..., Skinner [/bib_ref]. Once the agar pads containing cells were dried, glass coverslips (25 by 75 by 0.17 mm) were placed on the top of the cells. The slides with the coverslips were also taped at the corners to make them stable enough for the microscope imaging. Agar pads were maintained in an onstage incubator (catalog no. AMC1000; Thermo Fisher Scientific) in which temperature and humidity were controlled during resuscitation and microscope imaging. Both phase-contrast and fluorescence (GFP) images were obtained using a fluorescence microscope (Evos FL Auto 2; catalog no. AMAFD2000; Thermo Fisher Scientific) with a 100Â (oil) objective (Olympus, catalog no. AMEP4733; working distance, 0.3 mm) to determine cellular morphology and assess expression of the FtsZ-GFP fusion protein. An Evos GFP light cube (catalog no. AMEP4651) was used to acquire GFP images with 470/22-nm excitation and 510/42-nm emission wavelengths. To confirm that the pUA66-ftsZ-gfp plasmid can report the cellular ring structures, exponentially growing WT cells (harboring the reporter) that did not receive any treatment were monitored with microscopy as described above. The same microscopy procedure was also applied for the OFX-treated cells that were not starved.
(ii) Imaging cells harboring the pUA66-lon and pBAD-ftsZ-gfp plasmids. Stationary-phase WT or Dlon cells containing the pUA66-lon and pBAD-ftsZ-gfp plasmids were diluted 1:100 in 2 mL LB broth in 14-mL test tubes and then treated with OFX for 6 h. Arabinose (1 mM) was added to overnight and main cultures to induce ftsZ-gfp expression. OFX-treated cells were collected, washed with PBS solution, and resuspended in 2 mL LB, supplemented with 10 mM arabinose (61 mM IPTG) in 14-mL test tubes and grown in a shaker. At the indicated time points, one test tube was removed from the shaker; 1 mL of cell suspension was used for CFU enumeration, and the remaining 1 mL was used for the imaging process as described above.
(iii) SulA imaging. WT and Dlon cells containing the sulA reporter plasmid (pMSs201-P sulA -gfp) were collected after OFX treatment or UV exposure, serially diluted, spotted onto agarose pads, and imaged with microscopy, as described above. When necessary, cells treated with OFX or exposed to UV were transferred to 1 mL PBS solution in 5-mL test tubes and analyzed via flow cytometry to quantify GFPpositive (i.e., SulA-expressing) cells. Cells were excited at 470/20 nm, and green fluorescence was detected with a 510/42-nm emission filter. Cells carrying the EV were used as controls.
(iv) Image analysis. To capture a heterogenous cell population in each replicate, 10 different locations in each agarose pad, leading to a total number of cells ranging from 200 to 1,000, were selected and monitored with the use of the Evos FL Auto 2 imaging system. Cell morphology and FtsZ ring structures in microscope images were analyzed with ImageJ software [bib_ref] NIH Image to ImageJ: 25 years of image analysis, Schneider [/bib_ref]. Default ImageJ plugins and JavaScript (1.8.0) were used to process the images. The brightness of the phase-contrast images and the color of the fluorescent images were adjusted with "brightness/contrast," "color balance," and "sharpen" options. Given that dead, rough cells with aggregated proteins can be detected in both phase-contrast and fluorescent images, as shown in , we used phase-contrast images to investigate protein aggregation in the cells that do not have fluorescent protein expression systems. Cell lengths were measured manually using the segmented line tools in the ImageJ. The scale bars of raw images were used as a reference for the cell length measurements [bib_ref] NIH Image to ImageJ: 25 years of image analysis, Schneider [/bib_ref]. Similarly, the number of Z-rings within a bacterium was enumerated manually using the fluorescent images.
Statistical analysis. At least three independent biological replicates were performed for each experiment. A nonlinear logarithmic model was used to generate biphasic kill curves [bib_ref] Identifying metabolic inhibitors to reduce bacterial persistence, Mohiuddin [/bib_ref]. F statistics were used to compare kill curves and to assess statistical significance [bib_ref] Identifying metabolic inhibitors to reduce bacterial persistence, Mohiuddin [/bib_ref]. Pairwise comparisons were performed using oneway analysis of variance (ANOVA) with Dunnett's post hoc test or two-tailed t test with unequal variance. Each data point in the figures represents the mean value 6 standard deviation. For microscopy analyses, representative images from both phase-contrast and fluorescence microscopy are displayed in the figures. The threshold values for significance were set at *, P , 0.05; **, P , 0.01; ***, P , 0.001; ****, P , 0.0001; and ns, nonsignificant. GraphPad Prism was used to generate the figures and to perform the statistical analyses.
Binary logistic regression analysis from GraphPad Prism was used to determine whether the probability of resuscitation of a cell depends on the cell length (L), the number of Z-rings within the cell (Z) or the L/Z ratio. The outcomes (resuscitation status) are encoded as 1 (indicating a "success" in resuscitation) or 0 (indicating a "failure" in resuscitation). The form of the simple logistic model is expressed as follows:
log odds ð Þ¼ b 0 1b 1 X odds ¼ p 12p
where b 0 and b 1 are intercept and slope constants, respectively, and p is the probability of resuscitation of a cell. Wald test was used to determine if the slope of the simple logistic model (b 1 ) is significantly different from 0, which is equivalent to whether the odds ratio is 1.0 (i.e., P = 1.0).
# Supplemental material
Supplemental material is available online only.
[fig] FIG 1: Lon overexpression rescues colony formation in OFX-treated Dlon cells. (A and B) WT E. coli MG1655 (A) and Dlon E. coli MG1655 (B) [/fig]
[fig] FIG 3: Starvation in PBS solution rescues Dlon persisters. (A) WT and Dlon E. [/fig]
[fig] FIG 5: PBS starvation facilitates persister resuscitation in both WT and Dlon recovery cultures. (A) WT and Dlon containing an IPTG-inducible gfp expression plasmid (pUA66-gfp) were transferred to fresh LB recovery medium containing 1 mM IPTG after 6 h of OFX treatment. IPTG was not added to cultures before or during treatment. At designated time points during recovery, cells were collected and analyzed with a flow cytometer. (B) WT and Dlon cells containing pUA66-gfp were transferred to PBS solution after 6 h of OFX treatment and cultured at 37°C with shaking for 2 days. Cells were then collected and resuspended in fresh LB recovery medium and grown in the presence of 1 mM IPTG. At designated time points, cells were collected from the recovery culture and analyzed with a flow cytometer. (C and D) Cells described in panels A and B, respectively, were stained with PI at the indicated time point and analyzed with a flow cytometer. Live cells and ethanol (70% [vol/vol])-treated cells (dead cells) served as negative and positive controls, respectively. Proliferating cell fractions in the recovery populations are highlighted on the flow cytometry diagrams. A representative biological replicate is shown here. All replicates produced consistent results. n = 4. [/fig]
[fig] FIG 6: Starvation promotes Z-ring formation in OFX-treated Dlon cells. (A and B) WT (A) and the Dlon strain containing an IPTG-inducible FtsZ-GFP expression plasmid (pUA66-ftsZ-gfp) (B) [/fig]
[fig] FIG 7: The number and structural organization of Z-rings represent a key persister biomarker. WT and Dlon strains containing the IPTG-inducible FtsZ-GFP expression plasmid (pUA66-ftsZ-gfp) were transferred to PBS solution after 6 h of OFX treatment and cultured at 37°C with shaking. IPTG (1 mM) was added to cultures before and during OFX treatment to express FtsZ-GFP. Cells incubated in PBS solution for 2 days were then collected, pelleted by centrifugation, and resuspended in LB medium and were then spread on LB agarose pads containing 1 mM IPTG. The pads were monitored for 24 h with a fluorescence microscope with an onstage incubator. Cell lengths, as well as Z-ring structures and numbers, were determined for both resuscitating (persisters) and nonresuscitating (VBNC) cells to generate plots of the number of Z-rings versus cell length (A), the number of linear/spiral structures versus cell length (B), L/Z (cell length in mm/number of Z rings) ratios versus strain (C), L/Z versus cell length (D), and L/Z versus number of Z-rings (E). Pink color represents resuscitating cells, blue color represents nonresuscitating cells, and n represents the number of cells analyzed. [/fig]
[fig] FIG S1 ,: TIF file, 2.8 MB. FIG S2, TIF file, 2.7 MB. FIG S3, TIF file, 2.9 MB. FIG S4, TIF file, 2.6 MB. FIG S5, TIF file, 2.7 MB. FIG S6, TIF file, 2.8 MB. FIG S7, TIF file, 2.7 MB. FIG S8, TIF file, 2.9 MB. FIG S9, TIF file, 2.9 MB. TABLE S1, DOCX file, 0.02 MB. [/fig]
|
Interleukin 1 Polymorphisms, Lifestyle Factors, and Helicobacter pylori Infection
Associations between Helicobacter pylori (HP) infection and lifestyle factors have been reported by several authors, but little isknown about the host factors associated with the infection. This study aims to examine the infection rate of HP according to gene polymorphisms of interleukin (IL)-1A, IL-1B, and IL-1RN, and to investigate the interactions with lifestyle factors. Subjects were 241 non-cancer outpatients who had participated in a HP eradication program. Polymorphisms at − − − −889 (T to C) of IL-1A, at − − − −31 (C to T; T allele makes a TATA box) and − − − −511 (C to T) of IL-1B, and at intron 2 (86-bp VNTR (variable number of tandem repeats)) of IL-1RN were genotyped by PCR (polymerase chain reaction), PCR-RFLP (restriction fragment length polymorphism) and PCR-CTPP (PCR with confronting two-pair primers). It was found that IL-1B polymorphisms at − − − −31 and − − − −511 were near-completely linked, but in the opposite way to that in Caucasians; − − − −31C/ − − − −511T and − − − −31T/ − − − −511C alleles were dominant in the present subjects. The HP infection rate was substantially different among the genotypes of IL-1B C − − − −31T; 45.2% (19/42) for the C/C, 67.7% (90/133) for the C/T, and 63.6% (42/66) for the T/T. The age-sex adjusted odds ratio (OR) relative to the C/C genotype was 2.32 (95%CI (confidence interval), 1.10-4.92) for the T/C genotype and 2.46 (1.06-5.74) for the T/T genotype. The OR for the T/T genotype was significantly modified by smoking status; interaction term = = = =14.6 (1.12-190). The polymorphisms of IL-1A and IL-1RN were not associated with the infection rate. The results suggested that the T allele of IL-1B C − − − −31T is associated with vulnerability to persistent HP infection, and that the vulnerability is modified by smoking.
Helicobacter pylori (HP) infection, a well-known risk factor for stomach cancer, depends largely on sanitary conditions, especially in childhood. [bib_ref] The cohort effect and Helicobacter pylori, Banatvala [/bib_ref] [bib_ref] The transmission of Helicobacter pylori. A critical review of the evidence, Goodman [/bib_ref] There are still substantial differences in childhood prevalence between developed and developing countries, ranging from 4% to 82% reportedly. [bib_ref] The transmission of Helicobacter pylori. A critical review of the evidence, Goodman [/bib_ref] An increase in infected individuals with age is observed in many developed countries, which is considered to be due to a cohort effect. [bib_ref] The cohort effect and Helicobacter pylori, Banatvala [/bib_ref] Although there is no doubt that exposure to the bacterium increases the infection rate in the population, host susceptibility is also an important factor. Even after an ordinary level of HP exposure, uninfected persons still seem to exist in any population. In addition, spontaneous eradication after childhood infection may be influenced by host factors.
Polymorphisms of interleukin-1β gene (IL-1B) and interleukin 1 receptor antagonist gene (IL-1RN) were reportedly associated with stomach cancer risk for a population from Poland. 4) IL-1β is a pro-inflammatory cytokine with multiple biological effects, [bib_ref] Biologic basis for interleukin-1 in disease, Dianarello [/bib_ref] and is induced by HP infection. [bib_ref] Helicobacter pylori induces an array of pro-inflammatory cytokines in human gastric epithelial..., Jung [/bib_ref] IL-1B has three diallelic polymorphisms at −511, −31, and 3954 base pairs (bp) from the transcriptional start site. Though the differences were not statistically significant, persistent HP infection was observed for 60% of individuals harboring the C/C genotype (n=58) at −31 of IL-1B, for 73% of those with the C/T genotype (n=67), and for 79% of those with the T/T genotype (n=24) in a population from Scotland. [bib_ref] Interleukin-1 polymorphisms associated with increased risk of gastric cancer, El-Omar [/bib_ref] The T allele at −31 forms a TATA box, suspected of enhancing gene expression. The study reported that the polymorphism at −511 linked tightly with the polymorphism at −31, and therefore quite similar findings on the risks of stomach cancer and HP infection were obtained for IL-1B C−511T. The polymorphism at 3954 was not associated with the risks of stomach cancer and HP infection in the study. 4) IL-1RN has a penta-allelic 86-bp tandem repeat polymorphism in intron 2, which was found to affect IL-1β production in vitro, [bib_ref] Presence of the IL-1RA allele 2 (IL1RN * 2) is associated with..., Santtilla [/bib_ref] but it was not associated with HP infection. 4) IL-1α is another member of IL-1, which is encoded by IL-1A. It is primarily a cytosolic cytokine. About 10% or 15% is myristoylated and transported to the cell surface (membrane IL-1), but it is not detected in the serum under normal conditions. 5) IL-1A was reported to have three polymorphisms; C−889T, 46-bp variable number of tandem repeats (VNTR) at intron 6, and G4845T. A study showed that the combination of T/T of IL-1A C−889T and T/ − of IL-1B C−511T was related to high plasma levels of IL-1β. [bib_ref] A rare allele combination of the interluekin-1 gene complex is associated with..., Hulkkonen [/bib_ref] To date, no studies have been reported on the association of IL-1A polymorphisms with stomach cancer risk or HP infection.
The present case-control study examines the associations between persistent HP infection and four polymorphisms of the non-coding regions of IL-1; at −889 (T to C) of IL-1A, at −31 (C to T) and −511 (C to T) of IL-1B, and at intron 2 (86-bp VNTR) of IL-RN. To our knowledge, this is the first report on their genotype frequencies in Japanese. The interactions with lifestyle factors are also considered in this paper.
# Materials and methods
Study subjects Subjects were participants in an HP eradication program at Aichi Cancer Center Hospital, who were scheduled for gastroscopy. Written informed consent was obtained for a lifestyle questionnaire and gene polymorphism tests. The lifestyle questionnaire included smoking, alcohol, food intake frequency, physical exercise, and past history of cancer. In total, 283 outpatients (138 males and 145 females) participated in the study up to December 1999. We excluded 42 participants (38 with a history of cancer, 3 hepatitis virus carriers whose blood was not stored, and 1 who refused blood sampling after entry), and the remaining 241 outpatients were analyzed. HP infection An anti-HP IgG antibody test, High-molecular-weight Campylobacter-associated protein (HM-CAP) ELISA ("Detaminor H. pylori antibody," Enteric Products Inc., Westbury, NY) was used for detecting HP-infected
participants. An ELISA value of 2.3 or over was regarded as indicating HP infection-positive status. The sensitivity of HM-CAP was reported to be 98.7% and specificity 100% in the United States, [bib_ref] A sensitive and specific serologic test for detection of Campylobacter pylori infection, Evans [/bib_ref] though the sensitivity was not so high for Japanese. [bib_ref] Helicobacter pylori IgG antibody test established in the United States showed a..., Matsuo [/bib_ref] Genotyping DNA was extracted from 200 µl of buffy coat preserved at −40°C by the use of a QIAamp DNA Blood Mini Kit (QIAGEN Inc., Valencia, CA). The polymerase chain reaction (PCR) amplification was conducted using the primers listed in . The primers of IL-1RN were the same as described by and the others were chosen for this study. Genomic DNA (30 to 100 ng) was used in a volume of 25 µl with 0.1 mM dNTPs, 25 pmol of each primer, 0.5 units of "TaKaRa Taq" (TaKaRa Shuzo Co., Ltd., Otsu) or "AmpliTaq Gold" (Perkin-Elmer Corp., Foster City, CA), and 2.5 µl of 10× PCR buffer including 15 mM MgCl 2 . The PCR-RFLP (restriction fragment length polymorphism) method was used for the C-to-T single nucleotide polymorphism of IL-1B C−511T. The PCR products for tandem repeat polymorphisms of IL-1RN were used directly for electrophoresis. For the polymorphisms at −889 of IL-1A and at −31 of IL-1B, a new method named PCR-CTPP (PCR with confronting two-pair primers) was applied, which does not require a step to digest DNA products for single nucleotide polymorphism genotyping. 12) The primer setting for IL-1B C−31T is a revised version providing clearer bands than the former version. [bib_ref] Polymerase chain reaction with confronting two-pair primers for polymorphism genotyping, Hamajima [/bib_ref] [fig_ref] Figure 1: Gel showing the genotypes for polymorphisms at −889 of IL-1A, at −31... [/fig_ref] shows the results of PCR-CTPP for IL-1A C−889T and IL-1B C−31T, PCR-RFLP for IL-1B C−511T, and PCR for IL-1RN. All PCR products were visualized on a 2% agarose gel with ethidium bromide staining. To confirm genotyping for IL-1B C− 31T, PCR-RFLP was conducted using a restriction enzyme Underlined is the allele-specific base for PCR-CTPP.
(AluI) recognizing 5′-AGCT-3′, and DNA sequencing around −31 and −511 of IL-1B was performed for 4 samples.
Statistical analysis An unconditional logistic model was applied for estimating odds ratios (ORs) and interaction terms by a computer program, STATA Version 6 (STATA Corp., College Station, TX).
## Approval by ethical committee
This study had been approved by the Ethical Committee at Aichi Cancer Center in 1999 before the study started (Ethical Committee Approval Number 12-23).
# Results
The genotype frequencies are listed in [fig_ref] Table II: Distribution of Age and Genotypes of IL-1A, IL-1B, and IL-RN According to... [/fig_ref]. Forty subjects (16.6%) were found to harbor at least one T allele of IL-1A C−889T. A near-complete linkage disequilibrium opposite to that for Caucasians [bib_ref] Interleukin-1 polymorphisms associated with increased risk of gastric cancer, El-Omar [/bib_ref] was observed between polymorphisms at −31 and −511 of IL-1B; 17.2% (41/ 239) for C/C at −31 and T/T at −511, 54.0% (129/239) for C/T and C/T, 27.3% (65/239) for T/T and C/C, 1.3% (3/239) for C/T and C/C, 0.4% (1/239) for C/C and C/ C, and none for the other combinations. When all C/T at −31 and C/T at −511 were C-T and T-C alleles, only 5 out of 478 alleles (1.0%) were C-C allele and there was no T-T allele in the subjects. All genotypes of IL-1B C−31T determined by PCR-CTPP were the same as those by PCR-RFLP. DNA sequencing conducted for four samples (one T/T and C/C, one C/C and C/C, and two T/C and C/C) confirmed that the genotyping was correct. The participants harboring the 4/4 genotype of IL-1RN represented 90.0% of the total.
As shown in [fig_ref] Table II: Distribution of Age and Genotypes of IL-1A, IL-1B, and IL-RN According to... [/fig_ref] , the HP infection rate increased with age; this was especially clear for males. The rate was not different between the C/C genotype (61.7%) and C/T genotype (66.7%) of IL-1A C−889T (not significant), when males and females were combined. Those with the C/C genotype of IL-1B C−31T showed a lower infection rate (45.2%) than those with the C/T or T/T genotype (66.3%, χ 2 =5.62 with d.f.=1, P=0.018). Since the alleles of IL-1B C−511T are strongly linked with those of IL-1B C−31T, a corresponding finding was observed for the IL-1B C−511T polymorphism. No significant difference in the infection rate between the 4/4 genotype (62.2%, 135/ 217) and the others (66.7%, 16/24) of IL-1RN was observed.
Based on the above results, the ORs were calculated for the IL-1B C−31T genotype. The age-sex adjusted OR relative to the C/C genotype was 2.32 (95%CI (confidence interval), 1.10-4.92) for the C/T and 2.46 (1.06-5.74) for the T/T. [fig_ref] Table II: Distribution of Age and Genotypes of IL-1A, IL-1B, and IL-RN According to... [/fig_ref] shows the ORs according to lifestyle factor. The largest OR was observed for the T/T genotype among current smokers; OR=22.9 (1.97-266). The odds of antibody negative to positive were 10 to 6 for smokers with C/C, and 1 to 11 for smokers with T/T. The adjusted ORs of current smokers relative to non-current smokers were estimated for HP infection; 0.69 (0.14-3.27) for C/ C, 2.68 (0.86-8.32) for C/T, and 9.06 (0.97-84.1) for T/ T, indicating that smoking habit is a potent factor for persistent HP infection among individuals with the T/T genotype. When sex, age (a continuous variable), the C/T and T/T genotypes (two dummy variables), smoking (current vs. non-current), and interaction terms of smoking with the C/T and T/T genotypes (two variables) were included in a logistic model, the estimated interaction term between smoking and the T/T genotype was statistically significant; OR=14.6 (1. . This means that the effect of the T/T genotype for the current smokers was 14.6 times higher than that for the non-current smokers, or that the effect of current smoking for the T/T genotype was 14.6 times higher than that for the C/C genotype. Differences in OR were observed for milk intake (everyday vs. occasionally), fruits intake (≥4 times/week vs. less frequent), and physical exercise (≥3 times/week vs. less frequent), but the interaction terms with the genotype were not statistically significant. The subgroup analysis showed similar ORs for alcohol drinking (≥5 days/week and ≥1 gou or equivalent alcohol vs. other), salty seasoning (salty and average vs. not salty), intake frequencies of meat (≥4 times/week vs. less frequent), fish (≥4 times/week vs. less frequent), tofu (≥4 times/week vs. less frequent), fresh vegetables (≥4 times/week vs. less frequent), Japanese tea (≥6 cups/day vs. less), and coffee (everyday vs. occasionally).
# Discussion
Since anti-HP antibody tests have been most frequently used for epidemiologic studies on HP infection, [bib_ref] Epidemiology of, and risk factors for Helicobacter pylori infection among 2194 asymptomatic..., The [/bib_ref] [bib_ref] Helicobacter pylori infection and chronic atrophic gastritis among Japanese blood donors: a..., Fukao [/bib_ref] [bib_ref] Salty food intake and risk of Helicobacter pylori infection, Tsugane [/bib_ref] almost all epidemiologic findings on HP infection have been based on antibody tests. We also used an anti-HP antibody to determine persistent HP infection, although the test does not have 100% sensitivity for the existence of HP in the stomach. 10) Elevated antibody may reflect the response of the human body to HP, which is appropriate for the aim of investigating the association between disease risk of HP infection and genetic factors.
IL-1 is a multifunctional cytokine with high inflammatory activity. It has three members, IL-1α, IL-1β, and IL-1Ra, and there are two types of receptors, type I (IL-1RI) and type II (IL-1RII). Both receptors have a membranebound form and a soluble form. Only membrane-bound IL-1RI transduces a signal when combined with IL-1. IL-1RII is called a decoy receptor, because the membranebound form does not transduce a signal, nor does the solu- ble form. The five players (three IL-1 members and two receptors) regulate complex biologic reactions, in concert with other cytokines and small mediator molecules. 5) The polymorphisms of each player, whether known [bib_ref] Genetic polymorphism of human interleukin-1α, Bailly [/bib_ref] [bib_ref] The Danish Study Group of Diabetes in Childhood. Characterization of polymorphisms of..., Bergholdt [/bib_ref] or unknown, may affect IL-1 functions, independently and/or in combination. In this study, four polymorphisms were examined. Although the whole picture of their roles in the infection remains to be elucidated, the observed significant associations provide useful information to approach the host factors. For IL-1B, three polymorphisms, C−511T, C−31T, and C3954T, have been reported. A near-complete linkage disequilibrium between polymorphisms at −31 and −511 was found for the present participants. El-Omar et al. reported that the alleles at −31 were tightly linked with the alleles at −511; 0.610 for C-C combination, 0.010 for C-T, 0.005 for T-C, and 0.375 for T-T among the controls in Scotland (n=100), and 0.699 for C-C, 0.002 for C-T, 0.000 for T-C, and 0.298 for T-T among the controls in Poland (n=429). [bib_ref] Interleukin-1 polymorphisms associated with increased risk of gastric cancer, El-Omar [/bib_ref] In the present study, the corresponding values were 0.010, 0.438, 0.552, and 0.000, assuming that those with C/T at −31 and C/T at −511 did not harbor C-C or T-T allele. The sequence of IL-1B registered in GenBank (accession X04500) is the T-C type, very rare among Caucasians and common among our subjects. This complete discrepancy in the linkage within such a short distance as 480-bp between Caucasians and Japanese seems remarkable and should be further investigated. Since both studies showed that the T allele at −31 was related to a higher infection rate, though the effect was not significant in the El-Omar et al. study, the polymorphism at −511 seemed to be unrelated to the HP infection rate.
The T/T genotype at −31, a higher risk genotype for HP infection, was more frequent among our Japanese subjects (27.8%: 67/241) than among Caucasians (10.7%: 46/ 429 4) ). This suggests that Japanese are more vulnerable to persistent HP infection than Caucasians. The T allele of IL-1A was less frequent for our Japanese subjects (8.5%: 41/482) than for Caucasians (33.1%: 265/800 in Finland, [bib_ref] A rare allele combination of the interluekin-1 gene complex is associated with..., Hulkkonen [/bib_ref] and 28.3%: 137/484 in England 11) ). The alleles other than 4 repeats of IL-1RN were also rarer in the study participants (5.4%: 26/482) than in Caucasians (27.9%: 239/ 858, [bib_ref] Interleukin-1 polymorphisms associated with increased risk of gastric cancer, El-Omar [/bib_ref] 29.9%: 239/800 in Finland, [bib_ref] A rare allele combination of the interluekin-1 gene complex is associated with..., Hulkkonen [/bib_ref] and 26.6%: 139/522 in England 11) ). Although IL-1A, IL-1B, and IL-1RN are mapped within a 430 kilobase region on chromosome 2q, no clear linkages were observed among the polymorphisms except C−31T and C−511T of IL-1B in this study.
It was reported that 18 out of 59 patients with reflux esophagitis treated with omeprazole developed corpus atrophic gastritis, but none of 31 patients treated with fun- doplication did so,suggesting that acid secretion inhibition leads to HP re-distribution to the corpus and gastric atrophy, and possibly a condition favoring persistent infection. Since IL-1β is a strong inhibitor of gastric acid secretion, [bib_ref] Interleukin 1 beta and tumour necrosis factor alpha inhibit acid secretion in..., Beales [/bib_ref] individuals with the T/T genotype at −31, making a TATA box, may have a disadvantage in acquisition of HP infection and/or spontaneous eradication.
In the present study, lifestyle factors were found to modify the effect of IL-1B C−31T polymorphism on the HP infection. Some may be by chance due to multiple comparisons, while others may actually affect the association of the IL-1B polymorphism. Although further studies are needed to confirm the modification observed in this study, the findings are quite interesting in a biological sense. Smoking was reported to be associated with HP infection for inhabitants, [bib_ref] Epidemiology of Helicobacter pylori infection among 4742 randomly selected subjects from Northern..., Murray [/bib_ref] [bib_ref] An investigation into factors associated with Helicobacter pylori infection, Woodward [/bib_ref] outpatients, [bib_ref] Lifestyle and anti-Helicobacter pylori immunoglobulin G antibody among outpatients, Hamajima [/bib_ref] [bib_ref] Determinations of Helicobacter pylori infection and chronic gastritis, Fontham [/bib_ref] and participants in an HP eradication program, [bib_ref] Triple therapy with azithromycin, omeprazole, and amoxicillin is highly effective in the..., Bertoni [/bib_ref] while there are many studies reporting no association. [bib_ref] Epidemiology of, and risk factors for Helicobacter pylori infection among 2194 asymptomatic..., The [/bib_ref] [bib_ref] Salty food intake and risk of Helicobacter pylori infection, Tsugane [/bib_ref] Smoking interacted with the IL-1B C−31T polymorphism in this study. The infection rate was especially high among smokers harboring the T/T genotype, which is relatively rare for Caucasians. This difference in the genotype frequency may explain in part the above inconsistent results among the epidemiologic studies. Smoking is a well-known risk factor for gastric ulcer, which may suggest a synergistic effect with IL-1B activity for inflammation caused by HP infection, although there is no biological evidence for this.
Subgroup analysis for milk, fruits, and physical exercise showed different ORs, but the interaction was not signifi-cant. Accordingly, the difference could be caused by chance, while a small statistical power might fail to detect a significant interaction. Frequent fruits intake was reported to reduce the infection prevalence. [bib_ref] Determinations of Helicobacter pylori infection and chronic gastritis, Fontham [/bib_ref] This study demonstrated that the effect of IL-1B C−31T polymorphism was strong among those eating fruits 3 times or less per week. Salty food was reported to increase the HP infection rate, [bib_ref] Salty food intake and risk of Helicobacter pylori infection, Tsugane [/bib_ref] but it did not interact with this polymorphism. The other lifestyle factors, including alcohol drinking, also showed no interaction with the IL-1B C−31T polymorphism.
The interactions found in this study, significant or not, may provide a clue to resolve the biological mechanism of persistent HP infection. To confirm these interactions with lifestyle factors, as well as the association between this IL-1B polymorphism and HP infection, further studies will be required with different subjects.
[fig] Figure 1: Gel showing the genotypes for polymorphisms at −889 of IL-1A, at −31 and −511 of IL-1B, and at intron 2 of IL-1RN. Lane M contains a 100-bp DNA ladder. (a) IL-1A T−889C: lane 1 for C/C (279 and 428 bp), lane 2 for C/T (186, 279 and 428 bp), and lane 3 for T/T (186 and 428 bp), (b) IL-1B C−31T: lane 1 for C/C (345 and 574 bp), lane 2 for T/T (266 and 574 bp), and lane 3 for C/T (266, 345 and 574 bp), (c) IL-1B C−511T: lanes 8, 9 and 10 for C/C (194 bp) [/fig]
[table] Table II: Distribution of Age and Genotypes of IL-1A, IL-1B, and IL-RN According to Sex and Anti-HP Antibody Status [/table]
|
Genetic and environmental influences on quality of life: The COVID‐19 pandemic as a natural experiment
By treating the coronavirus disease 2019 (COVID-19) pandemic as a natural experiment, we examine the influence of substantial environmental change (i.e., lockdown measures) on individual differences in quality of life (QoL) in the Netherlands. We compare QoL scores before the pandemic (N = 25,772) to QoL scores during the pandemic (N = 17,222) in a sample of twins and their family members. On a 10-point scale, we find a significant decrease in mean QoL from 7.73 (SD = 1.06) before the pandemic to 7.02 (SD = 1.36) during the pandemic (Cohen's d = 0.49). Additionally, variance decomposition shows an increase in unique environmental variance during the pandemic (0.30-1.08), and a decrease in the heritability estimate from 30.9% to 15.5%.We hypothesize that the increased environmental variance is the result of lockdown measures not impacting everybody equally. Whether these effects persist over longer periods and how they impact health inequalities remain topics for future investigation.
# | introduction
Natural experiments pose a particularly interesting set of circumstances where an intervention is implemented that is not under the control of researchers. [bib_ref] Natural experiment methodology for research: a review of how different methods can..., Leatherdale [/bib_ref] With respect to research in the domain of public health and human behaviour, a great advantage of research on natural experiments is that it corresponds to 'real world' conditions, in contrast to many controlled experiments. Additionally, natural experiment studies are essential for evaluating population-scale (health) interventions and changes where experimental manipulation or random allocation is not feasible. As a result, natural experiments can provide unique ecologically valid insights into health processes as they are naturally occurring. A well-known example of a population-level natural experiment is the compulsory schooling age reform in the United Kingdom, where the minimum age at which students were allowed to leave school increased from 15 to 16 for everyone born on or after 1 September
1957. An interesting finding in the context of this reform is that the additional year of education reduced the gap in unhealthy body size between those in the top and bottom terciles of genetic risk for obesity from 20 to 6 percentage points, thus benefitting those with a higher genetic risk for obesity. [bib_ref] Education can reduce health differences related to genetic risk of obesity, Barcellos [/bib_ref] Another interesting set of natural experiments is the introduction of national tobacco control policies in different countries. For example, a workplace smoke-free legislation was introduced in Ireland in March 2004. One of the results of this legislation was an immediate significant reduction in small-for-gestational birth rates, which was sustained over the postban period. [bib_ref] Smoking ban and smallfor-gestational age births in Ireland, Kabir [/bib_ref] In the Netherlands, smoking prevalence decreased from 40%-51% to 22%-23% between 1993-1995 and 2009-2010, but no effect was seen on the heritability estimates of smoking. [bib_ref] Interplay between heritability of smoking and environmental conditions? A comparison of two..., Vink [/bib_ref] These examples involve national-level policy changes aimed at improving population health. Another set of natural experiments is (natural) disasters with population-level consequences. For example, on March 11, 2011, Japan was struck by an earthquake and consequent tsunami, leading to the loss of ±18,500 lives and ±345,000 people suffering damages to (or loss of) their house 5 and many people suffered from posttraumatic stress disorder (PTSD) after this disaster.
Hikichi and colleagues studied these events from a natural experiment perspective in order to gain knowledge on the association between social cohesion and the risk for PTSD. [bib_ref] Can community social cohesion prevent posttraumatic stress disorder in the aftermath of..., Hikichi [/bib_ref] They found that individualand community-level social cohesion before the disaster were associated with a lower risk of showing PTSD symptoms following the disaster. Another disastrous event with population-level consequences was World War 2 (WW2). During the horrific events of WW2, many children were separated from their parents. Pesonen and colleagues [bib_ref] Depressive symptoms in adults separated from their parents as children: a natural..., Pesonen [/bib_ref] March 2020, the World Health Organization (WHO) officially declared a pandemic, as the virus spread quickly across many countries in the world. As a result, many countries enforced a lockdown with varying levels of regulations. In the Netherlands, a so-called 'intelligent lockdown' was installed, meaning that public spaces, schools, restaurants, and so forth were closed and that people were encouraged to work from home, but could still leave their house for walks and other outdoor activities. As a result, many people's lives changed profoundly from an economic, social, and physical perspective.
What these different aspects (economic, social, physical) have in common is that they are all related to mental health and well-being. In a meta-analysis by Prati and Mancini, 7 the psychological impact of the COVID-19 pandemic lockdowns across 25 studies was evaluated in terms of both positive and negative psychological functioning. They found that lockdowns had a small but detrimental effect on mental health, as expressed in negative psychological functioning (i.e., anxiety, depression, substance use, sleep disturbances, suicide risk, negative affect, and general distress), but surprisingly the effects on positive psychological functioning were not significant. In a Dutch sample, specifically people without severe or chronic mental health disorders showed a slight increase in depression, anxiety, worry, and loneliness symptoms, whereas people with depressive, anxiety, or obsessive-compulsive disorders did not seem to have increased symptom severity during the pandemic compared with before. [bib_ref] The mental health impact of the COVID-19 pandemic on people with and..., Pan [/bib_ref] Besides the effects of this large natural experiment on mean population levels of mental health, such an impactful natural experiment enables a unique study into causes of individual differences in mental health.
From a behaviour genetic perspective, the focus goes beyond mean level changes to explain the causes of individual differences. It is well established that individual differences in well-being are influenced by both environmental factors and genetic factors:
research indicates that about 40% of individual differences in wellbeing is explained by genetic factors (the heritability), with the other 60% being explained by non-shared/unique environmental factors.Research combining behaviour genetics and experiments is relatively scarce, and typically focuses on short-term interventions. For example, one might use the 'method of co-twin control', where only one member of an identical twin pair receives an intervention. [bib_ref] Genetics and intervention research. Perspect, Plomin [/bib_ref] This is an interesting way of studying the possible effect of the intervention while controlling for genetic confounding. Alternatively, we can study individual differences in the effect of an intervention by applying an intervention in a classical twin design (CTD). This design also provides information on stability and change of the sources of individual differences pre-and post-intervention. For example, Haworth and colleagues examined the influence of a 10-week positive psychology intervention on well-being in a sample of 750 twins, and found that the relative influence of genetic and environmental influences remained stable, but that (partly) different non-shared environmental factors influenced well-being post-intervention. [bib_ref] Stability and change in genetic and environmental influences on well-being in response..., Haworth [/bib_ref] In a more recent study, a brief online mindset intervention increased the relative influence of additive genetic factors to individual differences in mindset. [bib_ref] Can a brief intervention alter genetic and environmental influences on psychological traits?..., Burgoyne [/bib_ref] The COVID-19 pandemic can serve as a natural experiment for the investigation of absolute and relative changes in the genetic and environmental causes of variation in well-being since we can compare the variance decomposition during the pandemic to before the pandemic.
For two well-being related constructs, optimism and meaning in life, it was already found that the heritability during the pandemic was slightly lower compared with before the pandemic.In addition to estimates of quantitative change such as lower heritability estimates, a study focusing on the qualitative aspects of the psychological responses to the COVID-19 crisis in young adults found a genetic correlation of 1 between pre-pandemic and pandemic purpose in life, indicating that the same genes affect this trait before and during the pandemic. 14 Optimism and meaning in life can be viewed as facets of well-being, [bib_ref] The construct validity of Ryff's scales of psychological well-being (SPWB) and their..., Kafka [/bib_ref] but whether these effects are similar for other wellbeing measures, such as quality of life (QoL), remains unexplored.
In the present study, we explore the impact of the COVID-19 pandemic on individual differences in well-being, quantified as QoL, in the Netherlands. We use a unique dataset that is comprised of data from twin families (e.g., twins, siblings, parents, aunts, uncles, nephews, nieces: pedigree data) both before and during the pandemic to provide a useful account of how genetic and environmental influences may be impacted by substantial environmental change.
## | materials and methods
## | participants
Participants were voluntary registrants of the Netherlands Twin Register (NTR). [bib_ref] The Netherlands twin register: longitudinal research based on twin and twin-family designs, Ligthart [/bib_ref] NTR participants are recruited through birth felicitation services, city councils, and online platforms. Every couple of years, biological and non-biological family members are invited to partake in survey research on development, health, behaviour, and lifestyle.
Relations among participants, that is, pedigree structure information, is stored in the 'Person Administration of the Netherlands Twin Register' (PANTER) database. [bib_ref] Design and implementation of a twin-family database for behavior genetics and genomics..., Boomsma [/bib_ref] Within this database, family roles and relations (e.g., mother-offspring, sibling-sibling) among participants are stored, with unlimited one-to-one relation possibilities for each individual. Participants can have multiple roles and relations in the database. For example, a person can be a mother and a twin.
For the current project, we selected a sample with pre-pandemic QoL data, and a (partly overlapping) sample with pandemic QoL data.
All participants were 16 years or older. For the pre-pandemic sample, QoL data were available for multiple waves of data collection. If multiple observations were available for an individual, we selected the most recent pre-pandemic observation (assessment data between January 2014 and February 2020). Within each family, if data for multiple siblings were available, we only selected data collected in the same data collection wave, in order to reduce potential timedependent confounders. Additionally, if data from both parent or spouses were available, we selected their data such that the data from both parents/spouses were included from the same wave of data collection.
During the pandemic, we made use of a single wave of data collection, which took place in April and May 2020, during the first lockdown in the Netherlands. Because we were interested in the effects of the lockdown on genetic and environmental influences on QoL, and not the effect of being infected itself, we excluded individuals with an (expected) COVID-19 infection (see below for details). We included twins and higher-order multiples (e.g., triplets), parents, siblings, and spouses (of multiples). Nuclear family information and age per type of family member is presented in . In total, prepandemic QoL data were available for 25,772 individuals, and pandemic QoL data were available for 17,222 individuals, of whom 11,232 had data available at both time points. Across the whole sample, age ranged between 16 and 102. Figures S1 and S2 visualize the pre-pandemic and pandemic age distributions, respectively.
## | measures
## | quality of life
Well-being was assessed as QoL using a Dutch version of Cantril's Self-Anchoring Striving Scale.Participants were asked the question:
'Where on the scale would you place your life in general? A score of 10 means the best life you can imagine, 1 means the worst life you can imagine'. In one of the pre-pandemic questionnaires, the question was scored on a scale from 0 to 10 instead of 1 to 10. As almost no participant scored a 0 (N = 6) or 1 (N = 3) on this question, these two answers were pooled together as one so that the question was scored similarly from 1 to 10 across the different questionnaires. [bib_ref] Using symptom-based case predictions to identify host genetic factors that contribute to..., Van Blokland [/bib_ref] [bib_ref] Real-time tracking of selfreported symptoms to predict potential COVID-19, Menni [/bib_ref] to predict whether a person likely had COVID at the time of assessment. Detailed information on the development and application of this variable can be found in the original study paper. [bib_ref] Real-time tracking of selfreported symptoms to predict potential COVID-19, Menni [/bib_ref]
## T a b l e 1 pedigree composition
## | statistical analyses
## | pre-pandemic to pandemic comparison
Means and standard deviations for pre-pandemic and pandemic QoL for individuals with different roles within families were calculated using R.We selected a subsample of genetically unrelated individuals (n = 8529) and performed a paired-samples t test to examine if QoL significantly changed from before to during the pandemic. Effect sizes were calculated using Cohen's d for paired samples. Additionally, we calculated within-individual difference scores that reflect individual change from before to during the pandemic.
## | kinship correlations
Kinship correlations were obtained to acquire a first indication of familial resemblance for QoL during and before the pandemic. We cal- were cross-time correlations. These last two correlations between pre-pandemic and pandemic QoL were pooled with fixed effect metaanalysis in the meta package in R so that one cross-phenotype correlation is computed to be used for further interpretation. As the tool does not provide standard errors or confidence intervals (CIs), these were calculated manually (s r ¼ ffiffiffiffiffiffiffi ffi 1Àr 2 nÀ2 q ).
## | genetic analyses
We used the Mendel 16.0 software package 'Variance Components'
analysis option [bib_ref] Mendel: the Swiss army knife of genetic analysis programs, Lange [/bib_ref] to decompose (co)variation in QoL into additive genetic (A), dominant genetic (D), common/household environmental (C), and unique environmental (E, also includes measurement error) sources of (co)variation. Effects of age and sex were regressed out prior to the Mendel analyses, and subsequent analyses were conducted on the residual QoL scores. [bib_ref] Unraveling the genetic and environmental relationship between well-being and depressive symptoms throughout..., Baselmans [/bib_ref] Shared environmental influences were defined as influences that are shared by members of the same household. As we are examining adults only, most (adult-aged) children within a nuclear family will not live in the same household.
Therefore, a household effect was specified for spouses.
To perform variance decomposition in Mendel, three input files are required: 1. an input pedigree file, 2. a control file, and 3. a definition file.
1. The input pedigree file contains all the familial and phenotype data, . 'sufficient' QoL before the pandemic (indicated by a 6 or higher), to 'insufficient' QoL during the pandemic (indicated by a 5 or lower).
## Figures 2 and 3 depict the number of individuals and percentage
of individuals, respectively, that increased, decreased, and remained stable per QoL pre-pandemic score. As can be seen in , prepandemic QoL scores are relatively skewed with most people indicating good pre-pandemic QoL. In general, the most common change was a decrease in QoL. Examining the group of respondents with decreased QoL during the pandemic in more detail , we see that individuals with high pre-pandemic QoL scores more often decreased during the pandemic compared with individuals with lower pre-pandemic QoL scores. With respect to the group of respondents that indicated increased QoL during the pandemic, it was especially individuals with lower pre-pandemic QoL scores that indicated higher scores during the pandemic. We also plotted the percentage of individuals that decreased, increased, or remained stable for QoL for different age groups separately in . Visual inspection of the plot does not show large differences between the age groups, with only a very slight trend of younger individuals being more negatively impacted in terms of QoL. Pandemic QoL correlations were similar to or lower than prepandemic QoL correlations. As seen in [fig_ref] Table 2: straints and consequently large environmental change in people's everyday lives, we did... [/fig_ref]
## | longitudinal and kinship correlations
## | genetic analyses
The total phenotypic variance in returned to more normal levels later on. It should be mentioned that some of the studies included in the reviews above also examined effects during the first lockdown. Therefore, it is likely that there are different effects across different countries, even during similar lockdown periods.
Important in the context of these results is that our sample, and the Netherlands in general, scores relatively high on QoL and other well-being measures compared with other countries. The Netherlands scores among the top happiest countries according to the 2019 World Happiness Report, [bib_ref] Changing world happiness, Helliwell [/bib_ref] which was unchanged in the 2021 World Happiness Report that reported on the data collected in 2020 (during the pandemic). [bib_ref] Happiness, trust, and deaths under COVID-19, Helliwell [/bib_ref] Importantly, we found a pandemic average QoL of 7.02 which is significantly lower than the pre-pandemic average, but still a good score indicating that people were still quite satisfied with their QoL. We found that especially those with higher QoL scores were prone to decreases in QoL during the pandemic. Given the skewed distribution of QoL in our sample, this was the majority of our sample.
Increases in QoL, however, were found mostly for individuals with lower QoL scores. While this is a relatively small part of our sample, it was surprising that it was especially individuals with lower baseline
QoL that showed improvements in QoL during the pandemic. A potential explanation is that we only examined individuals that did not have a COVID-19 infection around the time of assessment. Individuals with low levels of baseline QoL might have evaluated their QoL differently during the pandemic as they started comparing themselves with others that did become ill. In this way, they might have altered their perception, causing them to provide a different judgment during the pandemic. [bib_ref] The "eye of the hurricane" paradox: an unexpected and unequal rise of..., Recchi [/bib_ref] Individuals with higher levels of baseline QoL, on the other hand, might not have focused on these kinds of comparative mechanisms because they did not think their QoL was worse than average to begin with. Importantly, another possibility is that these findings might (partly) result from regression to the mean (RTM), the phenomenon whereby the second assessment of a trait results in values closer to the mean than at initial assessment purely by chance.
However, pre-pandemic QoL was measured on multiple occasions for some individuals, in which case we chose the latest available timepoint. By selecting participants from different measurement occasions, we attempt to acquire a better estimate of the participants' true baseline mean, which in turn decreases RTM. [bib_ref] Regression to the mean: what it is and how to deal with..., Barnett [/bib_ref] In a way, our prepandemic QoL measure is formed by taking a random sample around an individual's baseline mean levels. This makes it less likely that high/ low scorers will inevitably go down/up at the next measurement occasion (i.e., during the pandemic).Therefore, while we cannot rule out regression to the mean completely, this does make it less likely that our results are (fully) attributable to this phenomenon.
We used Mendel, instead of the CTD, to decompose the variance into genetic and environmental sources of variation. In the CTD, the variance components are estimated based only on the MZ and DZ twin covariances. As a result, only three parameters can be estimated simultaneously, so that an a priori choice needs to be made between an ACE or ADE model. The advantage of the Mendel software is that it allows for efficient analysis of whole pedigree data, allowing us to examine a large sample and estimating A, C, D, and E simultaneously.
There are extensions of twin designs where other family members can be included, such as the Cascade model [bib_ref] Modeling extended twin family data I: description of the cascade model, Keller [/bib_ref] that are more flexible in terms of model specification (e.g., constraining paths and sex-specific heritability). However, the advantage of Mendel is that it easily allows for the inclusion of complex family relations and irregular pedigrees, as are present in large twin-family registers like the Netherlands Twin
Register. Yet, we did not find any evidence for dominant genetic effects (D), that is, alleles acting in a multiplicative fashion (dominance or epistasis). Based on the correlations between the different types of family members ( It is important to interpret these results within the context of our sample and the time-frame in which we collected the data. Since it was the first lockdown, when the WHO had just announced a pandemic, individuals were likely still psychologically adjusting to the new situation. Whether the effects found in this study would be similar in later stages of the pandemic is a question that remains to be answered. Additionally, different countries employed different strategies to contain the virus, with the Netherlands installing the intelligent lockdown where people were encouraged to stay at home, but were still allowed to freely move around outside at all times of day. In this light, the finding of the large increase in environmental variance is even more remarkable, as the regulations in the Netherlands were less strict than those in many other countries. As such, the environmental effects of more stringent lockdown measures may be even larger. In any case, it is reasonable to expect that countries with different regulations will find different results than presented here, as these regulations impact the extent to which people had to alter their lives.
Finally, a limitation of our sample was that we had more female respondents than male respondents in both the pre-pandemic sample (65% female, 35% male), and the pandemic sample (71% female, 29% male). The representativeness was further limited by there being roughly twice as many highly educated individuals in the sample than expected based on the Dutch population.
# | conclusions
In conclusion, in this study we used data from before and during the
# Acknowledgments
We would like to thank all the twins and their family members for participation in the Netherlands Twin Register.
## Conflict of interest
The authors declare no relevant financial or non-financial interests to disclose.
## Ethics approval
All procedures performed in studies involving human participants
## Consent to participate
Written informed consent was obtained from all individual participants included in the study.
## Orcid
Margot P. van de Weijer https://orcid.org/0000-0001-9720-9481
[fig] |: grouped by family ID. Within the pedigree file, the following variables are required: family ID, person ID, and a Father and Mother ID, sex, and Twincode (an identifier for MZ twin pairs indicating which individuals are part of the same MZ twin pair). Genetic relationships between individuals within a pedigree with the same family ID are traced based on parental IDs (e.g., individuals within the same family with the same two parents are inferred to be full siblings). Our input pedigree file further specifies the household indicator field (in our case, spouse ID), and two fields for the (residualized) phenotype values for QoL before and during the pandemic. 2. The control file indicates all the analysis parameters. In our case, this includes the relevant variance components (A/C/D/E), the column names for the two quantitative traits present in the input pedigree file, the group factor specification (spousal household) and the way missing values are defined.3. Lastly, the definition file provides information on non-mandatory variables in the pedigree file: the variable types (factor/variable), and the associated levels and bounds. Mendel uses the control and definition files to read in the data from the input pedigree file, and estimates variance components based on variance-covariance matrices for relatives with different degrees of genetic relatedness based on classical biometrical genetics.24 We analysed four different models: 1) an ACDE model where C indicates the common household for spouses, 2) an ACE model, 3) an ADE model, and 4) an AE model. We compared the different nested models by comparing the log likelihood (LL) of the full ACDE model to the LL of the nested sub-models using a À2 log likelihood (À2LL) test that approximately follows a χ 2 distribution. Genetic and environmental correlations between the variance components were calculated by dividing the covariance of between pre-pandemic and pandemic QoL variables by the square root of its underlying variances. Pandemic to pre-pandemic comparison Mean QoL scores for the different groups can be found in [/fig]
[table] Table 2: straints and consequently large environmental change in people's everyday lives, we did not yet know whether this would lead to an increase or a decrease in environmental variance. Theoretically, the restrictions imposed by the lockdown measures could have reduced the environmental variance by making everyone's lives more similar to each other. However, the finding that these measures led to a large environmental increase suggests that the pandemic and consequent lockdown measures did not impact everybody in a similar way. It is important to identify such factors, since they potentially enlarge health inequalities during the pandemic. For example, a study by Ravens-Sieberer found that children with low socioeconomic status, migration background, and limited living space were affected significantly more by the pandemic in terms of health-related QoL, mental health problems, anxiety, and depression.38 Several potential explanations for individuals' different reactions to the environmental change imposed by the pandemic can be proposed. For example, people were encouraged to work from home, but only if possible. Before the pandemic, 1 in 3 people in the Netherlands (occasionally) worked from home. This increased to 1 in 2 people in the beginning of the pandemic, with the strongest increase in people with higher educational attainment and people using public transport to commute between home and work.39 Thus, the 'working from home' policy did not affect everyone in the population equally, potentially leading to increased differences in reported QoL across individuals. Additionally, with schools and day-care centres closed, people with children were likely impacted in a different way than people without children. Parents working from home (with children also staying at home) presumably had more trouble concentrating and were less productive, but the extent to which depended on different factors, like the age of the child and the age of parents.40 While the present study cannot pinpoint what exactly caused the increased environmental variance, these factors might serve assuggested causes of increased environmental variance in QoL in follow-up research. Based on the variance and covariance estimates provided by Mendel, we were able to calculate genetic and environmental correlations, which tells us something about the extent to which the same genetic and environmental factors influence QoL at the two time-points. We found a moderate genetic correlation (r A = 0.58) and common environmental correlation (r C = 0.40), and a small unique environmental correlation (r E = 0.05). To test if a correlation is significantly different from zero (or one), one would normally fit a model where the relevant correlation is constrained to zero (or one) and compare the fit of this model to the fit of the unconstrained model. Unfortunately, as Mendel does not allow for the inclusion of such constraints, we were not able to perform such model comparisons. However, based on the point estimate (E cov = 0.029) and standard error (SE = 0.031) of the unique environmental covariance, we can conclude that the unique environmental correlation is not significantly different from zero. In other words, the unique environmental factors influencing individual differences in QoL during the pandemic are completely different from the unique environmental factors influencing individual differences in QoL before the pandemic. This is important to consider when thinking about potential (positive) psychological prevention and intervention strategies to harness people from the negative effects of extreme environmental change, such as the COVID-19 pandemic lockdowns. These strategies rely on existing research by focusing their strategies on existing evidence on correlates of well-being. However, as indicated by this study, the environmental factors that determine individual differences in well-being under 'normal circumstances' are likely not the same as during crisis situations like these. [/table]
|
Lymphoepithelioma-like hepatocellular carcinoma
Rational: Lymphoepithelioma-like hepatocellular carcinoma (LEL-HCC) is a rare variant of hepatocellular carcinoma (HCC). To date, few cases have been reported in the literature, and almost no report in analyzing the different features of LEL-HCC.Patient concerns: We describe a 37-year-old female patient with a 32 Â 30 mm mass in the right liver.Interventions: Complete surgical resection of the lesion was performed.Diagnoses: Histopathological examination of the resected tumor revealed undifferentiated HCC cells with significant lymphocytes infiltration. Immunohistochemically, the tumor cells were positive for AFP (alpha fetoprotein), hepatocyte, CK8, and glypican-3. The patient was diagnosed with LEL-HCC.Outcomes: The patient had a favorable clinical outcome, and was free from tumor recurrence after a 52-months follow-up.Lessons: Our case was the youngest patient of all the reported cases, and the third case who was infected with both hepatitis B virus (HBV) and hepatitis C virus (HCV). LEL-HCC is a rare variant of HCC, with a relatively favorable prognosis. Further research recruiting more patients is required to determine the accurate causes and mechanism of LEL-HCC. Abbreviations: AFP = alpha fetoprotein, CT = computed tomography, EBV = Epstein-Barr virus, HBV = hepatitis B virus, HCC = hepatocellular carcinoma, HCV = hepatitis C virus, LELC = lymphoepithelioma-like carcinoma, LEL-CC = lymphoepitheliomalike cholangiocarcinoma, LEL-HCC = lymphoepithelioma-like hepatocellular carcinoma.
# Introduction
Hepatocellular carcinoma (HCC) is one of the most frequent malignancies worldwide, especially in developing countries; hepatitis B virus (HBV) and hepatitis C virus (HCV) infection are the main recognized risk factors. Despite of the progress of diagnostic and therapeutic approaches in recent years, the prognosis of HCC is still poor. It was reported that the overall 5year survival rate of HCC was approximately 5% to 6%. [bib_ref] The prognostic value of platelet count in patients with hepatocellular carcinoma: a..., Pang [/bib_ref] Lymphoepithelioma-like carcinoma (LELC) is one of the particular forms of undifferentiated epithelial carcinoma, characterized by massive lymphoid infiltration, and was first described in nasopharynx in 1982. [bib_ref] Lymphoepithelioma of the nasopharynx, Applebaum [/bib_ref] Subsequently, in various organs such as salivary glands, [bib_ref] Expression of Epstein-Barr virus in carcinomas of major salivary glands: a strong..., Tsai [/bib_ref] stomach, [bib_ref] Epstein-Barr virus associated lymphoepithelioma-like carcinoma at the lesser curvature of the upper..., Zhang [/bib_ref] lungs, [bib_ref] Lymphoepitheliomalike carcinoma of lung, Begin [/bib_ref] colon, [bib_ref] Lymphoepithelioma-like carcinoma of the colon, Mori [/bib_ref] thymus, [bib_ref] The role of Epstein-Barr virus in lymphoepithelioma-like carcinomas, Iezzoni [/bib_ref] uterine, [bib_ref] Lymphoepithelial-like carcinoma of the uterine cervix: a case report, Mori [/bib_ref] and ovaries, [bib_ref] Lymphoepithelioma-like carcinoma of the ovary: a case report and review of the..., Lee [/bib_ref] the particular tumor has been widely described. However, LELC is rarely reported in the liver. Hepatic LELC has 2 forms, lymphoepithelioma-like hepatocellular carcinoma (LEL-HCC) and lymphoepitheliomalike cholangiocarcinoma (LEL-CC). To date, at least 66 cases of LEL-HCC have been reported in the literature. [bib_ref] Lymphoepithelioma-like carcinoma in liver, Labgaa [/bib_ref] Herein, we reported another unique case of LEL-HCC with infection of both HBV and HCV, to further understand the rare disease's characteristics. A brief literature review of reported cases of LEL-HCC (including the present case) was also presented, which were summarized in [fig_ref] Table 1: Literature review of reported cases of LEL-HCC [/fig_ref]. [bib_ref] A long-term surviving patient with hepatocellular carcinoma including lymphocytes infiltration-a clinicopathological study, Shirabe [/bib_ref] [bib_ref] Clinicopathological study on hepatocellular carcinoma with lymphocytic infiltration, Wada [/bib_ref] [bib_ref] Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver..., Emile [/bib_ref] [bib_ref] Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity, Si [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma, Chen [/bib_ref] [bib_ref] Lymphoepitelioma-like hepatocellular carcinoma: a case report and a review of the literature, Nemolato [/bib_ref] [bib_ref] Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon, Park [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature, Shinoda [/bib_ref] [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] [bib_ref] Primary lymphoepitheliomalike hepatocellular carcinoma: report of a locally advanced case and review..., An [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: case report and review of the literature, Insilla [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: an uncommon variant of hepatocellular carcinoma with favorable outcome, Chan [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma without Epstein-Barr virus infection: a case report and a..., Wei [/bib_ref] [bib_ref] Programmed death ligand 1 expression in hepatocellular carcinoma: relationship with clinical and..., Calderaro [/bib_ref]
## Case presentation
A 37-year-old female patient with chronic HBV and HCV infection for decades was admitted to our hospital owing to a liver-occupying lesion during routine medical examination. Her past medical history was significant for diabetes mellitus and hypertension. Physical examination was unremarkable, except for mild abdominal tenderness. Abdominal contrast-enhanced computed tomography (CT) scan revealed a 32 mm  30 mm slightly hypodense nodular lesion in the right liver with obscure boundary [fig_ref] Figure 1: Abdominal CT showed a hypodense tumor [/fig_ref] ; arterial phase enhanced scan showed obvious enhancement [fig_ref] Figure 1: Abdominal CT showed a hypodense tumor [/fig_ref] and portal phase with mild heterogeneous enhancement [fig_ref] Figure 1: Abdominal CT showed a hypodense tumor [/fig_ref]. Pertinent laboratory investigations showed positive for HBV and HCV. The tumor Editor: Farid Azmoudeh-Ardalan.
J-KW and Y-WJ have equally contributed to the study and were the co-first authors.
The ethical approval from the Ethics Committee was not necessary to report this case because the case was reviewed retrospectively. Written informed consent was obtained from the patient for publication of this case report and all accompanying images. marker of alpha fetoprotein (AFP) was increased (105.90 ng/mL, normal <8). These findings were consistent with malignant nature of the liver tumor. Subsequently, a whole body positron emission tomography-computed tomography was performed, and there was no evidence of extrahepatic tumor. Based on above results, primary HCC was the initial diagnosis. The patient underwent surgery with complete tumor resection and the surgical margins were negative.
Macroscopically, the tumor was a yellowish solid mass, 32 mm in diameter. Histopathological examination of the resected specimen revealed that the tumor was consisted of undifferentiated HCC cells with significant lymphoid infiltration extending inside the tumor, which is the characteristic features of LELC. The epithelial tumor cells were featured by atypical, eosinophilic cytoplasm with large nuclei, prominent nucleoli. Immunohistochemically, the lymphoid infiltrate were composed of a mixture of CD8 and CD4 T-cells, with a predominance of CD8 cells. Meanwhile, the tumor cells were positive for AFP, hepatocyte, CK8, and glypican-3. No reactivity for CK7 or CK19 was observed. In situ hybridization for Epstein-Barr virus (EBV) was also negative. After discussion by several pathologists in our hospital, primary LEL-HCC was diagnosed. No postoperative chemotherapy or radiotherapy was performed for the patient. She had a favorable clinical outcome, and was free from tumor recurrence after a 52-months follow-up. Shirabe et al [bib_ref] A long-term surviving patient with hepatocellular carcinoma including lymphocytes infiltration-a clinicopathological study, Shirabe [/bib_ref] 58
[formula] /F N P N Y 1 2.2 R LR N N N N [/formula]
Alive without recurrence (83) Wada et al [bib_ref] Clinicopathological study on hepatocellular carcinoma with lymphocytic infiltration, Wada [/bib_ref] 62 * /10M, 1F N N P 6/11 U 2.2 * U LR N U U U 5-y survival: 100% Emile et al [bib_ref] Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver..., Emile [/bib_ref] 50
[formula] /M N P P Y 1 4.0 U LT N N N N Alive without recurrence (120) 54/M N N N N 2 2.0 U LT N Y Y N Died of disease (92) 59/M N P P Y 4 5.0 U LT N N N Y Alive without recurrence (96) 45/M N N P Y 1 2.0 U LT N N N Y Alive without recurrence (56) 64/M N N N Y 2 4.0 U LT N N Y Y [/formula]
Alive without recurrence (36) Si et al [bib_ref] Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity, Si [/bib_ref] 39 (5) Chen et al [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma, Chen [/bib_ref] 56 (21) Nemolato et al [bib_ref] Lymphoepitelioma-like hepatocellular carcinoma: a case report and a review of the literature, Nemolato [/bib_ref] 47
[formula] /F P N P Y 1 1.0 U LT N Y N N Died of disease/M N N P Y 1 3.0 R LR N Y N Y Died of disease/F N N N N 1 2.2 R LR N N N N [/formula]
Alive without recurrence (15) Park et al [bib_ref] Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon, Park [/bib_ref] 57
[formula] /M N P N Y 1 2.7 R LR N N N N [/formula]
Alive without recurrence (60) Shinoda et al [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature, Shinoda [/bib_ref] 79 (20) Patel et al [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] 74
[formula] /M N N N N 1 5.0 L LR Y Y N Y Alive with recurrence/F N N N N Multi 6.5 U LR N Y N U Died of disease (24) 65/M N N N N 1 4.8 U LR N N N U Died of unrelated cause (60) 65/F N N N N 1 1.3 U LR N N N U Alive without disease (108) 70/F N N N N 1 2.7 U LR N N N U Alive without recurrence (72) 61/F N N N N Multi 9.5 U LR N N N U Died of unrelated cause (4) 78/M N N N N 1 10.5 U LR N Y N U Alive with recurrence (48) 78/F N N N N 1 6.0 U LR N N N U Alive without recurrence (24) 57/F N N N N 1 13.0 U LR N N N U [/formula]
Died of unrelated cause (1) An et al [bib_ref] Primary lymphoepitheliomalike hepatocellular carcinoma: report of a locally advanced case and review..., An [/bib_ref] 50
[formula] /M N P N Y 1 3.0 R LR Y Y N Y [/formula]
Alive with recurrence (6) Cacciato et al [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: case report and review of the literature, Insilla [/bib_ref] 81/F N N P N 1 7.2 L LR N N N N Alive without recurrence (15) Chan et al [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: an uncommon variant of hepatocellular carcinoma with favorable outcome, Chan [/bib_ref] 58 (52) Calderaro et al [bib_ref] Programmed death ligand 1 expression in hepatocellular carcinoma: relationship with clinical and..., Calderaro [/bib_ref] mentioned 13 cases of LEL-HCC who underwent liver resection for hepatocellular carcinoma in a literature, but no relevant information was provided.
# Discussion
LELC is characterized by undifferentiated carcinoma cells with prominent infiltrating lymphocytes, originally described in nasopharynx. Subsequently, in various organs such as salivary glands, stomach, lungs, colon, thymus, uterine, and ovaries, the particular tumor has been widely described. In 2000, Emile et al retrospectively studied 162 cases of HCC who underwent orthotopic liver transplantation and found 5 cases who revealed abundant lymphoid stroma. [bib_ref] Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver..., Emile [/bib_ref] In the next year, Szekely proposed that "LELC" is defined as an undifferentiated epithelial tumor intermingled with a heavy lymphoid infiltrate. [bib_ref] Hepatocellular carcinoma with lymphoid stroma: "lymphoepithelioma-like carcinoma, Szekely [/bib_ref] In 2014, Patel et al reviewed all cases of HCC from 1988 in their institution and 8 cases of LEL-HCC were identified. [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] In 2015, Chan et al presented the largest series with 20 cases of LEL-HCC, from a 9-year retrospective cohort of patients who underwent surgical resection for HCC. [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: an uncommon variant of hepatocellular carcinoma with favorable outcome, Chan [/bib_ref] Recently, Labgaa et al reviewed the existing literature on LEL-HCC from epidemiological, clinicopathological, and research perspectives, and concluded that at least 66 cases of LEL-HCC have been reported till now. [bib_ref] Lymphoepithelioma-like carcinoma in liver, Labgaa [/bib_ref] The World Health Organization recognized this tumor as a variant of classical HCC.The clinicopathological features of LEL-HCC are not clear now. To further understand this rare disease's characteristics, we reviewed previous cases reported in the literature. Clinicopathological data of all the cases including our case were summarized in [fig_ref] Table 1: Literature review of reported cases of LEL-HCC [/fig_ref]. The male-to-female ratio was 1.7:1, and the mean age was 58 years (range, 37-81 years). Notably, our case was the youngest of all the patients. The cause of LEL-HCC remains unclear at present. EBV has been demonstrated to be strongly associated with LELC when the tumor exists in salivary glands, stomach, lung, and thymus. [bib_ref] The role of Epstein-Barr virus in lymphoepithelioma-like carcinomas, Iezzoni [/bib_ref] Also, the majority of LEL-CC are positive for EBV. [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] However, till now only one case of LEL-HCC was serology positive for EBV, [bib_ref] Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity, Si [/bib_ref] which may suggest that EBV infection is not a risk factor for LEL-HCC. In the only patient with EBV infection, he was dead of multiple recurrences 5 months after liver transplantation. Therefore, the biological implications of EBV in LEL-HCC need further discussion. HBV and HCV infection are widely recognized risk factors for HCC. The detail analysis of reported cases revealed that the rate of hepatitis virus infection in LEL-HCC was 70.4%, and the majority are HBV infection. Till now, only 3 patients were positive for both of HBV and HCV, [bib_ref] Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver..., Emile [/bib_ref] including the present case. Clinical presentations of LEL-HCC are nonspecific. Most patients were asymptomatic, and liver-occupying lesions were found during health examination. Some patients have right upper abdominal pain or symptoms like chronic cholangitis. [bib_ref] Lymphoepitelioma-like hepatocellular carcinoma: a case report and a review of the literature, Nemolato [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature, Shinoda [/bib_ref] Tumor marker of AFP was elevated in some cases, [bib_ref] Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon, Park [/bib_ref] which may provide a clue for the diagnosis of HCC. [fig_ref] Table 1: Literature review of reported cases of LEL-HCC [/fig_ref] showed that multiple nodular lesions were reported in 5 patients, whereas the others were single lesion. The mean tumor size was 3.8 cm, ranging from 1 to 13 cm. Notably, preoperative metastases were found in 2 cases with the presentation of enlarged lymph nodes next to the abdominal aorta, [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature, Shinoda [/bib_ref] [bib_ref] Primary lymphoepitheliomalike hepatocellular carcinoma: report of a locally advanced case and review..., An [/bib_ref] and radical resection was performed.
Definitive diagnosis of LEL-HCC can only be based on pathological examination, which often presents with poorly differentiated or undifferentiated large epithelial tumor cells with abundant lymphocytes infiltration. The tumor cells commonly show large nuclei and prominent nucleoli with abundant eosinophilic cytoplasm. The biological implications of the lymphocytes are currently under debate, which may be related to an antitumor effect. [bib_ref] Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon, Park [/bib_ref] Immunohistochemically, epithelial tumor cells are usually positive for various cytokines, such as CK8. Positive hepatocyte and glypican-3 indicated a hepatocellular origin rather than a cholangiocellular origin. Meanwhile, negative CK19 and CK7 also do not support a bile duct origin because they are biliary-type cytokeratins. In our case, the tumor cells were positive for CK8, glypican-3, AFP, and hepatocyte, but negative for CK7 or CK19, supporting the diagnosis of LEL-HCC. All reported cases showed a predominance of T-cells rather than B cells, except one case in which the cell dominancy was not available in the literature. [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] The T-cell markers varied among the previous reports. Patel et al reviewed 8 cases of LEL-HCC, and an equal distribution of CD4 and CD8 positive cells was observed. [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] Chan et al reported that the mean ratio of CD8 and CD4 was 6:1. [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: an uncommon variant of hepatocellular carcinoma with favorable outcome, Chan [/bib_ref] Cacciato et al observed a high predominance of T cells, and the majority were CD4 positive cells, with only a few CD8 positive cells at the margin of the lesion. [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: case report and review of the literature, Insilla [/bib_ref] The differential diagnosis included lymphoepithelioma-like cholangiocellular carcinoma and metastatic liver tumor. A whole body positron emission tomography-computed tomography excluding the possible existing extrahepatic disease supports the primary hepatic origin of the tumor.
The prognosis of LEL-HCC seems to be better than the conventional HCC in most cases, [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature, Shinoda [/bib_ref] [bib_ref] Characterization of inflammatory (lymphoepithelioma-like) hepatocellular carcinoma: a study of 8 cases, Patel [/bib_ref] [bib_ref] Primary lymphoepitheliomalike hepatocellular carcinoma: report of a locally advanced case and review..., An [/bib_ref] with a 5-year survival of 67% from [fig_ref] Table 1: Literature review of reported cases of LEL-HCC [/fig_ref]. However, the clinical outcomes were quite variable, ranging from aggressive clinical courses who died 5 months after surgery owing to recurrence, [bib_ref] Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity, Si [/bib_ref] to alive without recurrence for 10 years. [bib_ref] Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver..., Emile [/bib_ref] Interestingly, Park et al reported a case of regressing LEL-HCC, who was alive without recurrence for 50 months. [bib_ref] Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon, Park [/bib_ref] Postoperative recurrence was reported in 7 patients, and the shortest interval was 1 month after surgery. [bib_ref] Primary lymphoepitheliomalike hepatocellular carcinoma: report of a locally advanced case and review..., An [/bib_ref] In our case, 52 months have passed after surgery; the patient is still alive and free from tumor recurrence. Further accumulation and analyses recruiting more patients are needed to determine the prognosis of LEL-HCC.
There exists no consensus on standardized treatment strategy for LEL-HCC, and radical surgical resection or liver transplantation is still the most effective treatment currently. All of the patients with LEL-HCC, only 6, underwent liver transplantation, whereas the others underwent radical surgical resection. Some scholars have suggested that postoperative chemotherapy with sorafenib and cisplatin may benefit the patient [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma, Chen [/bib_ref] [bib_ref] Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature, Shinoda [/bib_ref] ; however, due to the rarity of this disease, the effectiveness of these chemotherapeutic substances has not been determined. Six patients received postoperative chemotherapy, of which 3 suffered tumor recurrence. In addition, immunosuppressive therapy may trigger tumor malignant progression, [bib_ref] Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity, Si [/bib_ref] which limited its use.
# Conclusions
LEL-HCC is a rare variant of HCC with a relatively favorable prognosis. We report a unique case of LEL-HCC with infection of both HBV and HCV. Further research recruiting more patients is needed to determine the accurate causes and mechanism of LEL-HCC.
[fig] Figure 1: Abdominal CT showed a hypodense tumor (red arrow) in the right liver: (A) plain scan phase; (B) arterial phase; and (C) portal venous phase. CT = computed tomography. [/fig]
[fig] F: = female, L = left liver, LEL-HCC = lymphoepithelioma-like hepatocellular carcinoma, LR = liver resection, LT = liver transplantation, M = male, Multi = multiple, N = no or negative, No. = number, P = positive, Post = postoperative, Pre = preoperative, R = right liver, U = unknown, Y = yes. * Indicated as mean value. [/fig]
[table] Table 1: Literature review of reported cases of LEL-HCC. [/table]
|
Modular evolution of secretion systems and virulence plasmids in a bacterial species complex
Background: Many named species as defined in current bacterial taxonomy correspond to species complexes. Uncertainties regarding the organization of their genetic diversity challenge research efforts. We utilized the Agrobacterium tumefaciens species complex (a.k.a. Agrobacterium biovar 1), a taxon known for its phytopathogenicity and applications in transformation, as a study system and devised strategies for investigating genome diversity and evolution of species complexes. Results: We utilized 35 genome assemblies, including 14 newly generated ones, to achieve a phylogenetically balanced sampling of A. tumefaciens. Our genomic analysis suggested that the 10 genomospecies described previously are distinct biological species and supported a quantitative guideline for species delineation. Furthermore, our inference of gene content and core-genome phylogeny allowed for investigations of genes critical in fitness and ecology. For the type VI secretion system (T6SS) involved in interbacterial competition and thought to be conserved, we detected multiple losses and one horizontal gene transfer. For the tumor-inducing plasmids (pTi) and pTi-encoded type IV secretion system (T4SS) that are essential for agrobacterial phytopathogenicity, we uncovered novel diversity and hypothesized their involvement in shaping this species complex. Intriguingly, for both T6SS and T4SS, genes encoding structural components are highly conserved, whereas extensive diversity exists for genes encoding effectors and other proteins.
# Conclusions:
We demonstrate that the combination of a phylogeny-guided sampling scheme and an emphasis on high-quality assemblies provides a cost-effective approach for robust analysis in evolutionary genomics. We show that the T6SS VgrG proteins involved in specific effector binding and delivery can be classified into distinct types based on domain organization. The co-occurrence patterns of VgrG-associated domains and the neighboring genes that encode different chaperones/effectors can be used to infer possible interacting partners. Similarly, the associations between plant host preference and the pTi type among these strains can be used to infer phenotypegenotype correspondence. Our strategies for multi-level investigations at scales that range from whole genomes to intragenic domains and phylogenetic depths from between-to within-species are applicable to other bacteria. Furthermore, modularity observed in the molecular evolution of genes and domains is useful for inferring functional constraints and informing experimental works.
Keywords: Agrobacterium, Genome, Secretion system, Virulence, Plasmid, Molecular evolution Background Understanding bacterial biology, notably for purposes of tackling pathogenicity, requires the ability to identify biological entities [bib_ref] The species concept for prokaryotes, Rosselló-Mora [/bib_ref] [bib_ref] The bacterial species challenge: making sense of genetic and ecological diversity, Fraser [/bib_ref] at the species level and to infer their evolutionary relationships. However, many bacterial groups are currently unresolved and are classified as species complexes. These uncertainties regarding species boundaries hamper research, communication, and policy-making such as in healthcare guidelines, pathogen quarantine regulations, and biological resource management. Based on barriers to homologous recombination, an analysis of > 20,000 bacterial genome sequences from 91 species belonging to 13 phyla found that 21 of the previously recognized species comprise multiple biological species [bib_ref] Biological species are universal across life's domains, Bobay [/bib_ref]. These 21 groups include those that are important as pathogens (e.g., Mycobacterium tuberculosis, Pseudomonas aeruginosa, and Vibrio cholerae) or beneficial microbes (e.g., Lactobacillus casei and Sinorhizobium meliloti). This finding highlights the ubiquity of species complexes across bacterial lineages, even for those that are extensively studied.
For such complexes, comprehensive understanding of the genetic diversity organization is required for robust species delineation, which in turn is essential for providing a reliable framework to interpret experimental findings and to gain insights into the biology. The use of genomic information has long been suggested as a powerful approach for defining species boundaries because the comprehensive genetic information can provide definitive and potentially quantitative guidelines [bib_ref] The bacterial species definition in the genomic era, Konstantinidis [/bib_ref]. However, several issues regarding genomic studies of bacterial species have remained unresolved. First, while genomospecies defined by overall genome divergence were suggested to represent distinct biological entities [bib_ref] Position taxonomique de souches de Agrobacterium d'origine hospitalière, Popoff [/bib_ref] [bib_ref] Homologous recombination in Agrobacterium: potential implications for the genomic species concept in..., Costechareyre [/bib_ref] [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] , the exact criteria for establishing the species boundaries are disputed. Although 95% average nucleotide identity (ANI) across the conserved parts of genomes was proposed as a universal boundary for defining species in bacteria [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref] , this criterion was challenged [bib_ref] Re-evaluating the evidence for a universal genetic boundary among microbial species, Murray [/bib_ref]. Additionally, ANI values alone do not provide information such as gene content or phylogenetic relationships, which are critical in understanding biological entities [bib_ref] The bacterial species definition in the genomic era, Konstantinidis [/bib_ref]. Second, phylogenetic relationships among closely related bacterial strains often cannot be resolved with confidence, yet such information is fundamental for evolutionary analysis. Third, comparative genomics studies are often limited by taxon sampling and/or assembly quality of available genome sequences.
In this study, we utilized the Agrobacterium tumefaciens species complex, also known as Agrobacterium biovar 1 [bib_ref] Agrobacterium-taxonomy of plant-pathogenic Rhizobium species, Young [/bib_ref] , as the study system for developing strategies that provide appropriate sampling and utilize multifaceted genomic analysis to resolve species boundaries and to investigate molecular evolution of key traits. These bacteria are known as the causative agents of crown gall disease that affects over 90 plant families [bib_ref] Historical account on gaining insights on the mechanism of crown gall tumorigenesis..., Kado [/bib_ref]. More importantly, the development of Agrobacterium-mediated transformation has provided a critical tool for genetic manipulation in plant sciences and agricultural biotechnology [bib_ref] Agrobacterium: nature's genetic engineer, Nester [/bib_ref]. Due to their importance, this complex has been studied for over a century and was found to harbor extensive phenotypic and genetic diversity that continues to confound efforts to resolve their taxonomy [bib_ref] Agrobacterium-taxonomy of plant-pathogenic Rhizobium species, Young [/bib_ref] [bib_ref] Historical account on gaining insights on the mechanism of crown gall tumorigenesis..., Kado [/bib_ref] [bib_ref] Agrobacterium: nature's genetic engineer, Nester [/bib_ref]. Various methods, such as DNA-DNA hybridization, biochemical characteristics, and molecular markers have been used to define 10 genomospecies (i.e., G1-G9 and G13), which have continually been associated to new nomenclature, a process that causes greater confusion than resolution [bib_ref] Position taxonomique de souches de Agrobacterium d'origine hospitalière, Popoff [/bib_ref] [bib_ref] Homologous recombination in Agrobacterium: potential implications for the genomic species concept in..., Costechareyre [/bib_ref] [bib_ref] A mathematical method for determining genome divergence and species delineation using AFLP, Mougel [/bib_ref] [bib_ref] Identification of genomic species in Agrobacterium biovar 1 by AFLP genomic markers, Portier [/bib_ref] [bib_ref] Rapid and efficient identification of Agrobacterium species by recA allele analysis: Agrobacterium..., Costechareyre [/bib_ref]. For example, the reference strain C58 used in many A. tumefaciens studies [bib_ref] A guide to Agrobacterium binary Ti vectors, Hellens [/bib_ref] [bib_ref] T-DNA binary vectors and systems, Lee [/bib_ref] belongs to G8, for which the name Agrobacterium fabrum was proposed in 2011 [bib_ref] Genomic species are ecological species as revealed by comparative genomics in Agrobacterium..., Lassalle [/bib_ref]. This has resulted in mixed usage of two names with different meanings (i.e., A. tumefaciens for the entire complex and A. fabrum for G8) in databases and literature. Compounding confusion, the name Agrobacterium radiobacter refers to A. tumefaciens G4 [bib_ref] Rapid and efficient identification of Agrobacterium species by recA allele analysis: Agrobacterium..., Costechareyre [/bib_ref] [bib_ref] Genomic species are ecological species as revealed by comparative genomics in Agrobacterium..., Lassalle [/bib_ref] and is also a heterotypic synonym of A. tumefaciens [bib_ref] Proposal that Agrobacterium radiobacter has priority over Agrobacterium tumefaciens. Request for an..., Young [/bib_ref]. Hereafter, we use A. tumefaciens in reference to the entire species complex and specific designations (i.e., G1-G9 and G13) for the genomospecies.
Previous characterizations found that A. tumefaciens strains have multipartite genomes with one circular chromosome, one linear chromosome, and highly variable plasmids [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Genome sequences of three Agrobacterium biovars help elucidate the evolution of multichromosome..., Slater [/bib_ref] [bib_ref] Genome sequence of the plant pathogen and biotechnology agent Agrobacterium tumefaciens C58, Goodner [/bib_ref] [bib_ref] The genome of the natural genetic engineer Agrobacterium tumefaciens C58, Wood [/bib_ref]. Consistent with the high levels of genetic divergence inferred from DNA-DNA hybridization [bib_ref] Position taxonomique de souches de Agrobacterium d'origine hospitalière, Popoff [/bib_ref] , cross-genomospecies comparisons typically found that < 80% of the genes are conserved [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Differentiations in gene content and expression response to virulence induction between two..., Haryono [/bib_ref]. Strikingly, > 32,000 horizontal gene transfer (HGT) events have been inferred to have shaped the evolutionary history of A. tumefaciens [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref]. Because the HGT patterns indicated co-transfers of genes that encode coherent biochemical pathways, it was hypothesized that purifying selection on those acquired gene clusters and overall gene content drove the ecological diversification among genomospecies [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref]. Moreover, the oncogenic plasmids that determine Agrobacterium phytopathogenicity exhibit complex modularity and transmission patterns, which further contributed to the diversification of these pathogens and their global spread [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref]. However, despite the progresses, those better-characterized genomospecies (e.g., G1, G4, G7, and G8) and pathogenic strains were highly overrepresented in previous genomics studies [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] , and such biases may affect our understanding of agrobacterial diversity and evolution.
To develop effective strategies for investigating bacterial species complexes such as A. tumefaciens, we started by performing targeted genome sequencing for strains in underrepresented lineages to achieve a balanced taxon sampling of the study system. The sampling scheme was based on information from two previous phylogenetic analyses of the A. tumefaciens species complex and its sister lineages, one based on recA [bib_ref] Rapid and efficient identification of Agrobacterium species by recA allele analysis: Agrobacterium..., Costechareyre [/bib_ref] and the other based on 24 conserved genes [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref]. We also limited analyses to only high-quality assemblies, which enabled detailed examinations of replicon-level synteny and confident inferences of gene presence/absence. The global view of genomic diversity and resolved phylogeny provided a robust framework for focused investigations of the genetic elements involved in key aspects of agrobacterial fitness and ecology, namely the type VI secretion system (T6SS) for interbacterial competition [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Agrobacterium tumefaciens deploys a superfamily of type VI secretion DNase effectors as..., Ma [/bib_ref] and the virulence plasmids for phytopathogenicity [bib_ref] Historical account on gaining insights on the mechanism of crown gall tumorigenesis..., Kado [/bib_ref] [bib_ref] Agrobacterium: nature's genetic engineer, Nester [/bib_ref]. Taken together, the investigations, scaling from whole-genome, wholereplicon, gene clusters, individual genes, and intragenic protein domains, provided novel and detailed information on the evolution and genetic diversity of bacteria important in plant pathology and biotechnology. Moreover, the strategies developed in this work are applicable to the study of other bacterial species complexes.
# Results
Genome sampling, molecular phylogeny, and divergence Based on existing knowledge of A. tumefaciens diversity [bib_ref] Position taxonomique de souches de Agrobacterium d'origine hospitalière, Popoff [/bib_ref] [bib_ref] Homologous recombination in Agrobacterium: potential implications for the genomic species concept in..., Costechareyre [/bib_ref] [bib_ref] A mathematical method for determining genome divergence and species delineation using AFLP, Mougel [/bib_ref] [bib_ref] Identification of genomic species in Agrobacterium biovar 1 by AFLP genomic markers, Portier [/bib_ref] [bib_ref] Rapid and efficient identification of Agrobacterium species by recA allele analysis: Agrobacterium..., Costechareyre [/bib_ref] and availability of genomic resources [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] , we identified 12 strains that represent six poorly characterized genomospecies [fig_ref] Table 1: List of the genome sequences used in this study [/fig_ref]. Additionally, two Agrobacterium larrymoorei strains, which represent the most closely-related sister lineage of A. tumefaciens [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Rapid and efficient identification of Agrobacterium species by recA allele analysis: Agrobacterium..., Costechareyre [/bib_ref] , were included as the outgroup. Whole-genome sequencing with substantial efforts in iterative improvements of the assemblies based on experimental and bioinformatic approaches were conducted for these 14 strains. Additionally, we selected 21 representatives from the 98 A. tumefaciens genome assemblies available from GenBank [fig_ref] Table 1: List of the genome sequences used in this study [/fig_ref] to yield a dataset with maximal genetic diversity without emphasis on including pathogenic strains. To ensure balanced sampling, we selected between two and five strains for each of the 10 recognized A. tumefaciens genomospecies. Importantly, 19 of these 35 assemblies, including nine produced in this study, are complete and most others are nearly complete (i.e., average N50 = 1.3 Mb; cf. the two chromosomes are~2.9 and~2.3 Mb, respectively).
Based on the homologous gene clustering results among these strains, we identified a core genome of 2093 singlecopy genes, which correspond to~40% of the genes annotated in each individual genome sequence. Compared to previous studies that conducted genome-based phylogenetic analysis for Agrobacterium or higher taxonomic ranks [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Taxonomy of rhizobia and agrobacteria from the Rhizobiaceae family in light of..., Ormeño-Orrillo [/bib_ref] , the more focused sampling in this study yielded a higher core gene count by one-to-two orders of magnitude. This increase in core gene count and the improvement in taxon sampling allowed for the inference of a well-resolved maximum likelihood phylogeny of the A. tumefaciens species complex [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. Each of the 10 currently recognized genomospecies forms a distinct monophyletic clade with > 80% bootstrap support. Additionally, we identified two novel genomospecies, G21 and G22, each represented by a single strain. The pattern of overall genome similarities exhibits a discrete multimodal distribution that supports use of a 95% ANI cutoff for delineating bacterial species [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref] and quantifies the divergence of the A. tumefaciens complex from its most closely related sister lineage [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref].
These 12 A. tumefaciens genomospecies were classified into seven groups based on 90% ANI and further assigned to three supergroups according to the phylogeny [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. Based on the time-calibrated phylogeny reported in Weisberg et al. [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] , the most recent common ancestor (MRCA) of A. tumefaciens emerged~48 million years ago (Mya) with a 95% highest posterior density (HPD) interval of 38.5-58.0 Mya, the three supergroups diverged~40 Mya (95% HPD interval 31.5-48.1 Mya), and most of the recognized genomospecies emerged~2-7 Mya (95% HPD interval 1.2-9.7 Mya). The large 95% HPD intervals suggest uncertainties regarding these time estimates. Regardless of the exact divergence time, the inferred rapid radiation in the early history of A. tumefaciens as shown by the short branch lengths likely prevented confident resolution of those deeper relationships in previous studies [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Taxonomy of rhizobia and agrobacteria from the Rhizobiaceae family in light of..., Ormeño-Orrillo [/bib_ref]. With improvements in the taxon sampling of this work, we observed~70% bootstrap support for those early nodes [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. This organismal tree provides a strong framework for our downstream examination of gene and domain phylogenies.
The high levels of assembly completeness provided confident inference of gene content comparisons. The principal coordinate analysis and hierarchical clustering results indicated that all 12 A. tumefaciens genomospecies Bootstrap support values in the range of 60-80% are labeled. Strains with complete genome assemblies are highlighted with an asterisk ("*"). The genomospecies assignments (i.e., G1-G9, G13, and G21-G22) are labeled to the right of strain names. The three A. tumefaciens supergroups are indicated by the colored background of the genomospecies assignments. Information to the right of the genomospecies assignments shows the grouping of genomes according to different cutoff values of genome-wide average nucleotide identity (ANI), the presence/absence of type VI secretion system (T6SS)-encoding gene cluster (green: present; white: absent), copy number of vgrG (white background: absent), number of plasmids, and the tumor-inducing plasmid (pTi) type based on k-mer profile (white background: absent). B Pairwise genome similarities based on the percentages of genomic segments mapped and the ANI values are similar to one another while distinct from A. larrymoorei (Additional file 1: [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] and S1C). Nonetheless, with the exception of G4 and G7, these genomospecies are distinguishable based on gene content (Additional file 1: [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] and S1D). This finding suggests that despite the extensive HGT inferred within this complex [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] , the genomospecies defined by 95% ANI likely represent distinct biological entities.
## Diversity and evolution of the t6ss genes
The confident inference of organismal phylogeny and gene content afforded by high-quality genome assemblies provided a robust framework for evolutionary analysis of key traits, particularly for those influence by the absence or loss of genes. We first focus on genes encoding the T6SS, a phage tail-like contractile nanomachine commonly found in Proteobacteria and used to inject effectors into eukaryotic or bacterial cells. The T6SS has major roles in pathogenesis, symbiosis, and interbacterial competition [bib_ref] Type VI secretion system effector proteins: effective weapons for bacterial competitiveness, Hernandez [/bib_ref] [bib_ref] Activity, delivery, and diversity of type VI secretion effectors, Jurėnas [/bib_ref] [bib_ref] The evolution of the type VI secretion system as a disintegration weapon, Smith [/bib_ref] [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref]. For A. tumefaciens, the T6SS is a key weapon for in planta competition between different genomospecies [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] and against other bacteria [bib_ref] Agrobacterium tumefaciens deploys a superfamily of type VI secretion DNase effectors as..., Ma [/bib_ref]. Thus, investigating the diversity and evolution of T6SS genes may shed light on a trait that influences the ecology and evolution of A. tumefaciens.
In a previous study that examined four A. tumefaciens genomospecies, T6SS-mediated anti-bacterial activity was observed for all 11 strains sampled and thought to be a conserved trait of this species complex [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref]. To our surprise, among the 33 A. tumefaciens strains examined in this work, a patchy distribution of the T6SS genes was observed [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. Gene absences are in strains corresponding to previously under-characterized genomospecies and were confirmed by examining syntenic regions and using TBLASTN [bib_ref] BLAST+: architecture and applications, Camacho [/bib_ref] to search entire genome sequences. For strains encoding a T6SS, corresponding genes are consistently located on the linear chromosome and mostly form a cluster of~20 genes organized as two adjacent and oppositely oriented imp and hcp operons [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Secretome analysis uncovers an Hcp-family protein secreted via a type VI secretion..., Wu [/bib_ref]. Some strains harbor accessory loci containing vgrG (involved in effector delivery) and other T6SS genes located elsewhere on the linear chromosome [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref] [bib_ref] Secretome analysis uncovers an Hcp-family protein secreted via a type VI secretion..., Wu [/bib_ref] (Additional file 1: . The T6SS gene phylogeny is largely congruent with the species tree . One notable exception is that the MRCA of G1 appears to have acquired the T6SS genes from a G8-related donor. Consistent with this inference, the T6SS genes in G1 strains are located in a different chromosomal location compared to strains of other genomospecies (Additional file 1: . Based on these observations, it is likely that a T6SS gene cluster was present in the MRCA of G8-G6-G14-G2-G4-G7-G9-G3-G15 and at least two independent losses have occurred in G6 and G14-G2. Regarding the ancestral state in the MRCA of the A. tumefaciens complex, presence of the T6SS genes appears to be a more parsimonious hypothesis based on the presence of these genes in the outgroup . However, the lack of synteny conservation between the linear chromosomes of A. tumefaciens and A. larrymoorei and the variable locations of vgrG homologs (Additional file 1: suggest that multiple independent origins are also possible. For broader scales, the T6SS genes have a patchy distribution among Rhizobiaceae [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref] [bib_ref] VgrG C terminus confers the type VI effector transport specificity and is..., Bondage [/bib_ref] , indicating that these genes are not essential for these bacteria and have high rates of gains and losses. Consistent with this, multiple pseudogenes confirmed by manual curation of annotation were found (e.g., tssH in S56 and tssA/vgrG in ATCC31749) , suggesting that for some strains these gene clusters are in the process of degradation and will be eventually lost.
Examination of synteny revealed that the imp operon, which encodes the majority of T6SS structural components [bib_ref] Systematic dissection of the Agrobacterium type VI secretion system reveals machinery and..., Lin [/bib_ref] , is more conserved in gene composition and order than the hcp operon, which often has different genes downstream of vgrG . This genetic diversity may play a key role in interbacterial competition because genes downstream of vgrG include those that encode effector and immunity (EI) protein pairs [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref]. The agrobacterial T6SS effectors often correspond to different toxins, and the cognate immunity proteins provide protection against self-intoxication [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Agrobacterium tumefaciens deploys a superfamily of type VI secretion DNase effectors as..., Ma [/bib_ref] [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref] [bib_ref] VgrG C terminus confers the type VI effector transport specificity and is..., Bondage [/bib_ref]. The rapid evolution of EI gene pairs is illustrated by three examples. First, despite the low levels of sequence divergence among G1 strains (i.e., > 98% ANI in all pairwise comparisons), different genes are found downstream of their vgrG homologs and this variation is not consistent with either the species phylogeny or the T6SS core gene phylogeny. Second, strain CFBP4996 of G7 has homologs of the same EI gene pair as strains 12D1 and 183 of G4, rather than with other members of G7. Third, in both G3 strains, vgrG and the associated EI genes are located elsewhere on the linear chromosome, rather than being a part of the hcp operon and Additional file 1: . These results suggest that recombination involving gene modules has contributed to the genetic diversity of these A. tumefaciens T6SS EI gene pairs.
## Modularity of vgrg and its associated ei pair
The knowledge that vgrG homologs encode proteins with distinct C-terminal domains responsible for binding specificities of different T6SS effectors for delivery suggested that each vgrG homolog and its downstream EI gene pair may evolve as a functional module [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref] [bib_ref] VgrG C terminus confers the type VI effector transport specificity and is..., Bondage [/bib_ref]. Here, we sought to investigate the patterns of gene co-occurrence and intra-module recombination to better understand the diversity and evolution of these genes. For in-depth investigation of vgrG evolution, we began by examining domain architecture of VgrG proteins and uncovered eight distinct domains (Additional file 1: . Based on differences in domain composition, the 44 vgrG homologs, including 17 associated with the main T6SS gene cluster and 27 associated with accessory loci, were classified into six major types and nine subtypes . Only three of the domains are present in all VgrG variants. The Nterminal domain 1 is the most conserved (Additional file 1: and the only one found in databases. This domain corresponds to TIGR03361, which accounts for6 6-77% of the protein length and the bulk of the structures that forms a trimeric complex analogous to a phage tail spike [bib_ref] Type VI secretion system translocates a phage tail spike-like protein into target..., Pukatzki [/bib_ref] [bib_ref] Type VI secretion apparatus and phage tail-associated protein complexes share a common..., Leiman [/bib_ref] (Additional file 1: [fig_ref] Figure 4: Gene neighborhoods of vgrG homologs [/fig_ref]. It is worth noting that the C-terminal end of domain 1 was identified as a recombination hotspot in a related study on agrobacterial T6SS genes [bib_ref] Diversification of the type VI secretion system in agrobacteria, Wu [/bib_ref]. For domain 5 that was found in all vgrG homologs belonging to subtypes A1-A3 and E1 , the presence of this domain is perfectly correlated with the presence of a downstream DUF4123-domaincontaining gene [fig_ref] Figure 4: Gene neighborhoods of vgrG homologs [/fig_ref]. Because this DUF4123 domain acts as an adaptor/chaperone for effector loading onto VgrG in A. tumefaciens [bib_ref] VgrG C terminus confers the type VI effector transport specificity and is..., Bondage [/bib_ref] and Vibrio cholerae [bib_ref] Identification of divergent type VI secretion effectors using a conserved chaperone domain, Liang [/bib_ref] [bib_ref] Chimeric adaptor proteins translocate diverse type VI secretion system effectors in Vibrio..., Unterweger [/bib_ref] , this strong co-occurrence suggests specific interactions between VgrG domain 5 and DUF4123. Thus, combining domain analysis with gene co-occurrence provides a new strategy for predicting the interacting domains of VgrG with other T6SS components.
Intriguingly, despite conservation of domain architecture within each subtype , the phylogenies inferred from the three domains encoded by all vgrG homologs do not have the same topology (Additional file 1: [fig_ref] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes [/fig_ref]. For domain 1, sequences from the same subtype do not always form monophyletic groups. For domains 2 and 3, the short sequence lengths limited the phylogenetic resolution; nonetheless, low divergence within the same subtype and high divergence between subtypes were observed. These patterns suggest that each domain-encoding region evolved independently and can recombine between subtypes.
At the level of gene cluster organization, vgrG homologs within a subtype can have distinct downstream genes (e.g., A1 and B1), regardless of whether they are associated with the main T6SS gene cluster [fig_ref] Figure 4: Gene neighborhoods of vgrG homologs [/fig_ref] and Additional file 2: Dataset S1). These findings suggest that in addition to domain shuffling among vgrG homologs [bib_ref] Diversification of the type VI secretion system in agrobacteria, Wu [/bib_ref] , recombination also facilitated novel vgrG-effector pairings in the evolution of these T6SS genes.
When T6SS diversity was examined in a phylogenetic context, numbers, and types of vgrG homologs, as well as their linked EI genes, lack strong correlations with species phylogeny [fig_ref] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes [/fig_ref]. Based on our manual curation of vgrG-associated genes, a total of 63 putative effector genes were identified (Additional file 2: Dataset S2). Among these, peptidoglycan-targeting toxins and nucleases are the two most commonly found categories with Phylogeny and organization of the T6SS gene cluster. The maximum likelihood phylogeny was inferred based on a concatenated alignment (5960 aligned amino acid sites) of 14 core T6SS genes, including tagE, tagF, tssM, tssL, tssK, tagH, tssG, tssF, tssE, tagJ, tssC40, tssC41, tssB, and tssD. Two other core genes, tssA and tssH, are excluded because some homologs are pseudogenized. Genes downstream of tssD (e.g., tai, tae, and vgrG) are excluded due to variable presence. The species phylogeny on the right is based on [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. Genes are color-coded according to annotation, and syntenic regions are indicated by gray blocks Domain organization and classification of vgrG homologs. Six major types with two types having subtypes (A1-3 and D1-2) within them were identified and labeled on the right. For each homolog, the genomospecies assignment is provided in square brackets, followed by the locus tag. The gene names are provided in parenthesis for those functionally characterized homologs (i.e., vgrG1-2 for C58 homologs and vgrGa-d for 1D1609 homologs) 21 each. This finding is consistent with an investigation of > 1000 T6SS effectors sampled from 466 species in the phylum Proteobacteria [bib_ref] The evolution of the type VI secretion system as a disintegration weapon, Smith [/bib_ref] , which also found these two as the dominant categories and suggested that they play complementary roles in T6SS-mediated competition. Based on our homologous gene clustering results, these 63 putative effector genes were classified into 20 families [fig_ref] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes [/fig_ref]. Among these, 11 were experimentally validated in previous studies, including EI1/6/15 in Ma et al. [bib_ref] Agrobacterium tumefaciens deploys a superfamily of type VI secretion DNase effectors as..., Ma [/bib_ref] , EI4/11 in Wu et al. [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] , EI7 in Santos et al. [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref] , and EI2/5/8/9/10 in Wu et al. [bib_ref] Diversification of the type VI secretion system in agrobacteria, Wu [/bib_ref].
## The virulence plasmids and associated genes
The tumor-inducing plasmids (pTi) are an important component of A. tumefaciens genomes. These large accessory replicons harbor the virulence (vir) regulon genes that encode the Vir proteins and type IV secretion system (T4SS) for processing and delivering a transfer . For each vgrG homolog, three upstream genes are plotted to illustrate if it is associated with the main T6SS gene cluster or not, and 10 downstream genes are plotted to illustrate putative effector/immunity genes. Two A1-type homologs from the strain S56 (i.e., AGR1B_RS27940 and AGR1B_RS27960) are in close association with each other and plotted together DNA (T-DNA) into plant cells, and are essential for agrobacterial phytopathogenicity [bib_ref] Historical account on gaining insights on the mechanism of crown gall tumorigenesis..., Kado [/bib_ref] [bib_ref] Agrobacterium: nature's genetic engineer, Nester [/bib_ref]. Among the 35 strains examined, we identified 15 pTi sequences . Two novel putative pTi (i.e., pTiCFBP4996 and pTiCFBP5473) were found in the 14 newly sequenced strains. This efficiency of discovering novel pTi types is surprising, given our previous study that defined pTi types I-VI was based on extensive sampling of diverse historical collections containing 162 Agrobacterium strains [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref]. This finding demonstrates the importance and usefulness of a phylogeny-guided approach for investigating genetic diversity. Our recent examination of > 4000 Rhizobiaceae plasmids sampled from 1251 strains representing 222 species-level taxa assigned these two novel putative pTi to types X and XI [bib_ref] Diversification of plasmids in a genus of pathogenic and nitrogen-fixing bacteria, Weisberg [/bib_ref]. Both types are rare; type X is found in CFBP5473 and only one other A. larrymoorei strain (AF3.44), CFBP4996 is the only strain that harbors a type XI plasmid. These two novel pTi are distinctive in their large sizes , gene organization [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref] , gene content (Additional file 1: [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref] , and sequence divergence of core genes (Additional file 1: [fig_ref] Figure 7: Distribution of key vir genes among the putative pTi [/fig_ref]. Moreover, their T-DNA regions also differ from the typical sizes of~18-26 kb observed in types I-III pTi [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. For the tumorigenic strain CFBP5473 (Additional file 1: , the predicted T-DNA border sequences flank an exceptionally large (~93 kb) region that contains all of the vir regulon genes in addition to the typical T-DNA-associated genes (e.g., synthesis of opine and plant hormone) [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. This reflects either a translocation of a T-DNA border sequence or reliance on a non-canonical T-DNA border sequence that we could not identify. For pTiCFBP4996, its 7-kb T-DNA is predicted to contain only four genes (i.e., two correspond to opine synthesis and two encode hypothetical proteins). Plant hormone synthesis genes, which are necessary to cause visible disease symptoms, were not identified in this predicted T-DNA region or elsewhere on this plasmid. Consistent with predictions, strain CFBP4996 did not induce tumor formation when inoculated onto stems of tomato plants (Additional file 1: . Considering that pTiCFBP4996 encodes all essential vir genes required for T-DNA processing and transfer, CFBP4996 may serve as a naturally disarmed strain capable of T-DNA transfer without causing diseases.
For replicon-level comparisons, types II and III pTi are more similar to each other than to type I pTi based on gene content (Additional file 1: [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref] and core gene phylogeny (Additional file 1: [fig_ref] Figure 7: Distribution of key vir genes among the putative pTi [/fig_ref]. Within type I, the two subtypes (i.e., I.a and I.b) are distinguishable by gene content (Additional file 1: [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref] but do not form mutually exclusive clades in the core-gene phylogeny (Additional file 1: [fig_ref] Figure 7: Distribution of key vir genes among the putative pTi [/fig_ref]. All putative pTi, including the two novel types and pTiS4 (i.e., type V from distantly-related Agrobacterium vitis), contain genes for the T4SS that mediates T-DNA transfer into plant cells (i.e., virB1-B11 and virD4) and the corresponding twocomponent regulatory system (i.e., virA and virG) [fig_ref] Figure 7: Distribution of key vir genes among the putative pTi [/fig_ref]. Strain 1D1609 is a notable case because its virA and virJ are located on another plasmid, rather than the pTi [bib_ref] Differentiations in gene content and expression response to virulence induction between two..., Haryono [/bib_ref]. For the other vir regulon genes, several differences among pTi types were observed [fig_ref] Figure 7: Distribution of key vir genes among the putative pTi [/fig_ref]. For example, while all pTi harbor a conserved virE3a that facilitates T-DNA protection and entry into host [bib_ref] Agrobacterium delivers anchorage protein VirE3 for companion VirE2 to aggregate at host..., Li [/bib_ref] , type I pTi harbor one or two additional copies of virE3 that belong to different sequence types (i.e., sharing the same annotation but classified as distinct homologs due to sequence divergence). Similarly, virF, which encodes an Fbox protein that is a putative host-range determinant [bib_ref] virF, the host-range-determining virulence gene of Agrobacterium tumefaciens, affects T-DNA transfer to..., Jarchow [/bib_ref] , can be classified into three sequence types with distinct distributions. Those less well-characterized vir genes, such as virD3 [bib_ref] The Agrobacterium tumefaciens virD3 gene is not essential for tumorigenicity on plants, Vogel [/bib_ref] [bib_ref] The virD4 gene is required for virulence while virD3 and orf5 are..., Lin [/bib_ref] , virJ [bib_ref] An Agrobacterium virulence factor encoded by a Ti plasmid gene or a..., Pan [/bib_ref] , and virP, are also distributed differently among pTi types. Finally, in addition to the presence/absence of individual genes, the overall organization of vir regulons also differ among these pTi (Additional file 1: . All type I pTi are conserved in sharing a~40-kb region that contains all vir genes. In comparison, locations of vir genes are more variable among type II/III pTi; virF/P (and virQ/H if present) are located~5-50 kb away from the main vir gene cluster. Other than vir genes, the gene content and organization of T-DNA are also different . All List of the pTi sequences analyzed. These include 15 from the genome data set listed in [fig_ref] Table 1: List of the genome sequences used in this study [/fig_ref] [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. Within the T-DNA regions, the plant hormone synthesis genes (i.e., tms1/iaaM, tms2/iaaH, and ipt) are the most conserved ones, while others are more variable . Taken together, this genetic variation may contribute to the host range differences observed among strains harboring different types of pTi. For example, based on a comparison among > 100 strains, there are strong associations between type I and III pTi with woody and herbaceous plants, respectively [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref]. Moreover, previous tumorigenesis assays demonstrated that strains harboring types I and II pTi tend to have higher virulence against Brassicaceae and Asteraceae hosts, respectively [bib_ref] Characterization and host range of five tumorigenic Agrobacterium tumefaciens strains and possible..., Hwang [/bib_ref].
# Discussion
## Biological entities at the species level and above
Based on the divergence of core gene sequences [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] and gene content (Additional file 1: [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] , 95% ANI is a reliable approach for defining species within the A. tumefaciens complex, as is the case for most other bacteria [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref]. The discrete multimodal distribution of genome similarities [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] suggested that there are genetic barriers between different genomospecies, which may be explained by neutral processes and/or selection [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref]. Regardless of the exact mechanisms, these patterns supported application of the biological species concept, which is based on genetic barriers, to these A. tumefaciens genomospecies and other bacteria [bib_ref] Biological species are universal across life's domains, Bobay [/bib_ref]. With the continuing drop in sequencing cost, ANI analysis can serve as a standard approach for accurate classification of additional strains [bib_ref] Minimal standards for the description of new genera and species of rhizobia..., De Lajudie [/bib_ref] , which in turn could facilitate research and communication, and ideally leads to improvements in bacterial taxonomy for basic works and applications.
It is worth noting that despite our effort, the 33 strains included in this study do not fully capture the diversity of the A. tumefaciens species complex. For example, the species Agrobacterium arsenijevicii [bib_ref] Agrobacterium arsenijevicii sp. nov., isolated from crown gall tumors on raspberry and..., Kuzmanović [/bib_ref] , Agrobacterium nepotum (G14) [bib_ref] Revised phylogeny of Rhizobiaceae: proposal of the delineation of Pararhizobium gen. nov.,..., Mousavi [/bib_ref] , Agrobacterium viscosum (G15) [bib_ref] Two novel genomospecies in the Agrobacterium tumefaciens species complex associated with rose..., Mafakheri [/bib_ref] , and two unnamed genomospecies G19/G20 [bib_ref] Two novel genomospecies in the Agrobacterium tumefaciens species complex associated with rose..., Mafakheri [/bib_ref] are also members of this species complex. As more strains are characterized in the future, it is likely that higher levels of diversity within this group will continue to be discovered. Furthermore, several potentially confusing issues regarding Agrobacterium taxonomy remain to be resolved. For example, strain MAFF210266 that we referred to as a representative of G21 shares 98% ANI with an important strain K599 (=NCPPB2659) [bib_ref] Draft genome sequence of Agrobacterium rhizogenes strain NCPPB2659, Valdes Franco [/bib_ref]. However, K599 harbors a root-inducing plasmid (pRi) associated with hairy root disease and was named as [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. The species phylogeny on the right is based on [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. The pTi sequences derived from this study are highlighted in bold. For those with relevant information available, the genomospecies assignments are indicated in square brackets and the opine types are indicated in parentheses. For the alignment, all plasmids are visualized in linear form starting from the replication genes. The key gene clusters are color-coded according to functions, and the predicted T-DNA regions are indicated by the black horizontal bracket above each plasmid. Syntenic regions are indicated by gray blocks Agrobacterium rhizogenes (a.k.a. Agrobacterium biovar 2 [bib_ref] Agrobacterium-taxonomy of plant-pathogenic Rhizobium species, Young [/bib_ref] or G10 [bib_ref] A mathematical method for determining genome divergence and species delineation using AFLP, Mougel [/bib_ref] , all deprecated synonyms for Rhizobium rhizogenes), which is phylogenetically divergent from the A. tumefaciens species complex. Recently, during the revision of this work, the name Agrobacterium G21 was independently proposed and strain K599 was reclassified to G21 [bib_ref] Comparative genomics of novel Agrobacterium G3 strains isolated from the International Space..., Singh [/bib_ref].This example highlights the dynamic nature of pTi/pRi transmission within the agrobacteria-rhizobia complex [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] and advocates a classification scheme based on genome-wide ANI, rather than plasmid content or phenotype.
At the above-species level, A. tumefaciens genomospecies exhibited some intriguing patterns of genome divergence. In a previous study that compared~90,000 prokaryotic genomes, it was extremely rare to find ANI values in the range of 82-96% [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref]. In other words, strains either belong to the same natural biological entity at the species level and have > 95% ANI, or belong to different species and have < 82% ANI. This observation could be due to biases in the sampling of available genomes, or the all-against-all pairwise comparisons included mostly distantly-related species [bib_ref] Re-evaluating the evidence for a universal genetic boundary among microbial species, Murray [/bib_ref]. In our study that provided a detailed examination of closely related species, the~85-93% ANI among A. tumefaciens genomospecies [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] indicated that the A. tumefaciens complex is indeed a coherent entity with high divergence from its closest sister lineage within the same genus (i.e., A. larrymoorei). The driving forces for maintaining species complexes and the prevalence of such above-species level entities are interesting questions that require further investigations. For the A. tumefaciens complex, although the nomenclature originated from its phytopathogenicity, it is well-established that this group contains both pathogenic and non-pathogenic strains that differ in the possession of an oncogenic plasmid (i.e., pTi) or not [bib_ref] Agrobacterium-taxonomy of plant-pathogenic Rhizobium species, Young [/bib_ref]. The promiscuous nature of their pTi [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Transfer of the Agrobacterium tumefaciens Ti plasmid to avirulent agrobacteria and to..., Hooykaas [/bib_ref] [bib_ref] Presence of an Agrobacterium-type tumor-inducing plasmid in Neorhizobium sp. NCHU2750 and the..., Haryono [/bib_ref] [bib_ref] Alternative non-Agrobacterium based methods for plant transformation, Rathore [/bib_ref] [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref] suggested that lineages within this complex may experience frequent transitions between Gene presence and absence are indicated by filled and empty circles, respectively. The main vir genes include the components of the type IV secretion system (T4SS; virB1-B11 and virD4). For virE3, the three sequence types are listed separately. For virF and virD3, the sequence types are labeled inside the circles. For 1D1609, the genes virA and virJ are located on another plasmid and plotted as absent in this figure. The locus tags are provided in Supplementary Dataset S1B pathogenic and non-pathogenic lifestyles, and such shared ecological niches may be the force that maintains the coherence of this species complex. Compared to sister lineages (e.g., A. larrymoorei and A. rubi), the more diverse host range of A. tumefaciens [bib_ref] Agrobacterium-taxonomy of plant-pathogenic Rhizobium species, Young [/bib_ref] [bib_ref] Historical account on gaining insights on the mechanism of crown gall tumorigenesis..., Kado [/bib_ref] may be linked to the higher diversity of pTi types [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] , which may have facilitated the divergence of this complex into multiple genomospecies. To test this hypothesis, better sampling of these sister lineages is required.
At the genus level and above, genome-based classification is more challenging. The ANI approach is expected to have limited resolution when nucleotide sequence identities are below~80% [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref]. Moreover, the fractions of genome sequences alignable for ANI value calculation are highly variable for genus-level comparisons [bib_ref] A genus definition for Bacteria and Archaea based on a standard genome..., Barco [/bib_ref] , which raises concerns on the robustness of applying the ANI method to higher taxonomic ranks. To resolve this challenge, analysis of protein sequence divergence among core genes was proposed as a suitable approach [bib_ref] A standardized bacterial taxonomy based on genome phylogeny substantially revises the tree..., Parks [/bib_ref]. However, while it may be desirable to establish a standardized taxonomy with a defined range of genomic divergence for each taxonomic rank, large variations in the divergence values at a given rank were observed among different taxonomic groups in previous attempts [bib_ref] A genus definition for Bacteria and Archaea based on a standard genome..., Barco [/bib_ref] [bib_ref] A standardized bacterial taxonomy based on genome phylogeny substantially revises the tree..., Parks [/bib_ref] [bib_ref] A complete domain-to-species taxonomy for Bacteria and Archaea, Parks [/bib_ref]. These variations created situations where some families contain higher divergence levels than some orders or lower divergence levels than some genera, even after normalization and a full revision of the current taxonomy [bib_ref] A standardized bacterial taxonomy based on genome phylogeny substantially revises the tree..., Parks [/bib_ref]. Such situations demonstrated the challenges of establishing a new standardized taxonomy even when the practical issues of transitioning from the current taxonomy are not considered, and perhaps is to be expected given the highly variable evolutionary rates across different lineages [bib_ref] Inferring clocks when lacking rocks: the variable rates of molecular evolution in..., Kuo [/bib_ref]. Given these considerations, other aspects of biology (e.g., physiology, ecology) may play more important roles in defining those higher taxonomic ranks.
## Units and modularity of molecular evolution
For evolutionary studies, the levels at which selection and other processes operate on have been a topic that received much attention. For prokaryotes, levels from the entire genome to individual functional domains within genes are of particular interest. Based on our results, all of these levels must be considered to comprehend the complex patterns. At the whole-genome level, the clear species boundaries based on overall similarity [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] and Additional file 1: [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref] suggested that the entire genome largely evolves as a single coherent unit. This result is consistent with previous findings that at the global level HGT has very little impact on the reconstruction of organismal phylogeny [bib_ref] Phylogenetics and the cohesion of bacterial genomes, Daubin [/bib_ref] [bib_ref] Global extent of horizontal gene transfer, Choi [/bib_ref] , despite the extensive HGT inferred in bacterial evolution [bib_ref] Global extent of horizontal gene transfer, Choi [/bib_ref] [bib_ref] Lateral gene transfer and the nature of bacterial innovation, Ochman [/bib_ref] [bib_ref] Modular networks and cumulative impact of lateral transfer in prokaryote genome evolution, Dagan [/bib_ref] [bib_ref] Lateral transfer of genes and gene fragments in prokaryotes, Chan [/bib_ref] and the importance of HGT in adaptation [bib_ref] Adaptive evolution of bacterial metabolic networks by horizontal gene transfer, Pál [/bib_ref] [bib_ref] The fate of new bacterial genes, Kuo [/bib_ref] [bib_ref] Origins of bacterial diversity through horizontal genetic transfer and adaptation to new..., Wiedenbeck [/bib_ref]. A possible explanation for these seemingly conflicting observations is that most Organization of the transfer DNA (T-DNA) on tumor-inducing plasmids (pTi). Two unusual putative pTi sequences (i.e., pTiCFBP4996 and pTiCFBP5473) are excluded, and other pTi sequences derived from this study are highlighted in bold. For those with relevant information available, the genomospecies assignments are indicated in square brackets and the opine types are indicated in parentheses. Genes are colorcoded according to annotation, syntenic regions are indicated by gray blocks of the acquired genes are lost quickly [bib_ref] The fate of new bacterial genes, Kuo [/bib_ref] , presumably due to the strong mutational bias towards deletions observed in bacteria [bib_ref] Deletional bias and the evolution of bacterial genomes, Mira [/bib_ref]. Additionally, acquired genes are subjected to the selection that drives species diversification [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref] , which is expected to act on all genes in a genome together.
At the level of individual replicons, chromosomes and plasmids certainly have distinct evolutionary histories [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. Because novel chromosome/plasmid combinations may lead to speciation [bib_ref] Presence of an Agrobacterium-type tumor-inducing plasmid in Neorhizobium sp. NCHU2750 and the..., Haryono [/bib_ref] , and the spread of plasmids has important implications on the evolution of virulence [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Genomic insights into the contribution of phytopathogenic bacterial plasmids to the evolutionary..., Sundin [/bib_ref] and antimicrobial resistance [bib_ref] Plasmid encoded antibiotic resistance: acquisition and transfer of antibiotic resistance genes in..., Bennett [/bib_ref] , further investigations on the evolution of plasmids and their compatibilities with chromosomes are important [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] [bib_ref] Pathways for horizontal gene transfer in bacteria revealed by a global map..., Redondo-Salvo [/bib_ref]. Additionally, for bacteria with multiple chromosomes, examining the evolution of individual chromosomes may provide novel insights. In the case of A. tumefaciens, the multipartite genome was hypothesized to originate from intragenomic gene transfer from the ancestral circular chromosome to a plasmid, followed by linearization of this plasmid to form the secondary chromosome [bib_ref] Genome sequences of three Agrobacterium biovars help elucidate the evolution of multichromosome..., Slater [/bib_ref] [bib_ref] Single acquisition of protelomerase gave rise to speciation of a large and..., Ramírez-Bahena [/bib_ref]. The secondary chromosome is known to exhibit higher levels of divergence in overall organization, gene content, and sequences [bib_ref] Genome sequences of three Agrobacterium biovars help elucidate the evolution of multichromosome..., Slater [/bib_ref] [bib_ref] Differentiations in gene content and expression response to virulence induction between two..., Haryono [/bib_ref]. In this regard, it is curious to note that the apparently rapidevolving T6SS genes are all located on the secondary chromosome, rather than the more conserved primary chromosome. For future studies, it may be interesting to compare the molecular evolution of T6SS and other genes between species with mono-and multi-partite genomes.
At the levels of gene clusters and below, several interesting observations were made based on the loci of the two secretion systems investigated in this work. First, although these two systems may provide some fitness advantages (e.g., T6SS for interbacterial competitions and T4SS for host exploitation), complex patterns of gains and losses were observed [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. These patterns suggest that there is not a strong selective pressure to maintain these genes and non-adaptive stochastic processes are important. Alternatively, these genes may be subjected to heterogeneous selective pressures. Regardless, there is a certain degree of modularity regarding their evolution, such that the presence patterns are all-or-none and no partial cluster was found for the chromosomal T6SS genes or the plasmid-encoded T4SS genes. This is particularly evident for the T6SS genes, as when the main cluster was lost (e.g., G2 and G6), no accessory vgrG loci located elsewhere was found (Additional file 1: . Second, for both systems, genes for the structural components are conserved, while those for the effectors and others are not [fig_ref] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes [/fig_ref]. This is consistent with the expectation that opposite selective forces may act on these two categories of genes, with purifying selection against changes to preserve the apparatus of a functional secretion system and positive selection for more diverse effectors and other accessory components. Third, modularity that may reflect functional constraints were observed at finer scales. For example, each vgrG is linked to its cognate effector/chaperone genes, and each effector is linked to its cognate immunity gene. Such modularity is expected to be maintained by selection, similar to the observations regarding co-transfers of genes involved in associated biochemical pathways [bib_ref] Ancestral genome estimation reveals the history of ecological diversification in Agrobacterium, Lassalle [/bib_ref]. However, the linkage could be broken down by recombination at within-or between-species levels, as evident in the diversity of vgrG gene neighborhoods, even for those homologs belonging to the same type [fig_ref] Figure 4: Gene neighborhoods of vgrG homologs [/fig_ref]. Fourth, at the level of individual genes, the within-genome diversity of vgrG homologs [fig_ref] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes [/fig_ref] provides further support to the hypothesis that HGT is more important than duplications in driving gene family expansions in bacteria [bib_ref] Horizontal transfer, not duplication, drives the expansion of protein families in prokaryotes, Treangen [/bib_ref]. Finally, at the intra-gene level, within-or between-species recombination may be important in generating novel combinations of domains, thus promoting the diversification of homologs and Additional file 1: [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref].
Taken together, these observations illustrated the complexity of biological systems. While it is difficult to draw up generalized rules or to estimate the relative importance of each evolutionary process at different levels, it is important to consider and examine these complexities to better understand organisms of interest.
# Conclusions
In summary, by using a group of important bacteria as the study system, this work utilized a strategy of phylogenyconscious genome sampling for systematic investigations of a species complex. This approach requires prior knowledge regarding the extant phylogenetic diversity of the study system, which is important in the planning stage of large-scale genome-sequencing projects. In addition to improving the cost-effectiveness, this strategy is also critical in obtaining an unbiased picture that does not overemphasize certain subgroups. Furthermore, the emphasis on generating and utilizing high-quality assemblies improves the confidence in gene content analysis. With the continuing advancements in sequencing technologies and bioinformatic tools, this emphasis becomes increasingly accessible. For this study system, our examination of biological boundaries at the species level and above improves the understanding of how natural biodiversity is organized. The targeted analysis of those secretion system genes and oncogenic plasmids provides novel insights regarding the key genetic variations involved in the fitness and ecology of these soil-borne phytopathogens that need to compete in complex microbiota and invade plant hosts. Moreover, the multi-level analysis of their genetic diversity from whole-genome to intra-genic domains highlights the complexity of these biological systems. The strategy and findings of this work provide useful guides for future studies of other bacteria.
# Methods
## Genome sequencing
A total of 14 strains were acquired from the French International Center for Microbial Resources (CIRM) Collection for Plant-associated Bacteria (CFBP) [fig_ref] Table 1: List of the genome sequences used in this study [/fig_ref]. These include 12 strains that belong to the A. tumefaciens species complex and two A. larrymoorei strains as the outgroup.
The procedure for whole-genome shotgun sequencing was based on that described in our previous studies [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Complete genome sequence of Agrobacterium tumefaciens Ach5, Huang [/bib_ref] [bib_ref] Complete genome sequence of Agrobacterium tumefaciens 1D1609, Cho [/bib_ref]. All bioinformatics tools were used with the default settings unless stated otherwise. Briefly, total genomic DNA was prepared using the Wizard Genomic DNA purification kit (Promega, USA). The Illumina paired-end sequencing libraries were prepared using KAPA LTP Library Preparation Kits (Roche Sequencing, USA) with a targeted insert size of~550 bp. The Illumina MiSeq platform was used to generate 300 × 2 reads with an average coverage of 306-fold per strain (range 141-to 443-fold). The raw reads were quality trimmed using a Q20 cutoff and used for de novo assembly based on Velvet v1.2.10 [bib_ref] Velvet: algorithms for de novo short read assembly using de Bruijn graphs, Zerbino [/bib_ref] with the settings "-exp_cov auto -min_contig_lgth 2000 -scaffolding no." The contigs were oriented by mapping to those complete genome assemblies available [fig_ref] Table 1: List of the genome sequences used in this study [/fig_ref] using MAUVE v2015-02-13 [bib_ref] Mauve: multiple alignment of conserved genomic sequence with rearrangements, Darling [/bib_ref]. Due to the difficulties of identifying appropriate reference genomes for several evolutionary branches, four strains (i.e., CFBP5473, CFBP5875, CFBP5877, and CFBP6623) were selected for PacBio long-read sequencing and PacBio HGAP v3 assembly. These PacBiobased assemblies were used as a guide for scaffolding, rather than the finalized results.
To improve the Illumina-based draft assemblies, an iterative process was used to examine the raw reads mapping results and to incorporate gap-filling results based on PCR and Sanger sequencing. This process was repeated until the complete assembly was obtained or the draft assembly could not be improved further. The finalized assemblies were submitted to the National Center for Biotechnology Information (NCBI) and annotated using the Prokaryotic Genome Annotation Pipeline (PGAP) [bib_ref] NCBI prokaryotic genome annotation pipeline, Tatusova [/bib_ref].
## Comparative and evolutionary analysis
The genomes analyzed are listed in [fig_ref] Table 1: List of the genome sequences used in this study [/fig_ref]. The procedures for genome comparisons were based on those described in our previous studies [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Comparative genome analysis of Spiroplasma melliferum IPMB4A, a honeybee-associated bacterium, Lo [/bib_ref] [bib_ref] Species boundaries and molecular markers for the classification of 16SrI phytoplasmas inferred..., Cho [/bib_ref]. Briefly, pairwise genome similarities were calculated using Fas-tANI v1.1 [bib_ref] High throughput ANI analysis of 90 K prokaryotic genomes reveals clear species..., Jain [/bib_ref]. For comparisons of plasmids, FastANI was executed with the custom settings that reduced fragment length to 1000 bp and minimum matched fragments to [bib_ref] The genome of the natural genetic engineer Agrobacterium tumefaciens C58, Wood [/bib_ref]. For global alignments of chromosomes and plasmids, the syntenic regions were identified by BLASTN v2.6.0 [bib_ref] BLAST+: architecture and applications, Camacho [/bib_ref] and visualized using genoPlotR v0.8.9 [bib_ref] genoPlotR: comparative gene and genome visualization in R, Guy [/bib_ref]. For gene content comparison, BLASTP v2.6.0 [bib_ref] BLAST+: architecture and applications, Camacho [/bib_ref] with e-value cutoff set to 1e −15 and OrthoMCL v1.3 [bib_ref] OrthoMCL: identification of ortholog groups for eukaryotic genomes, Li [/bib_ref] were used to infer the homologous gene clusters. The result was converted into a matrix of 35 genomes by 17,058 clusters, with the value in each cell corresponding to the copy number. This matrix was further converted into a Jaccard distance matrix among genomes using the VEGAN package v2.5-6 in R, then processed using the principal coordinates analysis function in the APE package [bib_ref] ape 3.0: New tools for distance-based phylogenetics and evolutionary analysis in R, Popescu [/bib_ref] and visualized using ggplot2 v3.3.2. The hierarchical clustering analysis was performed using PVCLUST v3.4.4 [bib_ref] Pvclust: an R package for assessing the uncertainty in hierarchical clustering, Suzuki [/bib_ref].
For phylogenetic analysis, homologous sequences were aligned using MUSCLE v3.8.31 [bib_ref] MUSCLE: multiple sequence alignment with high accuracy and high throughput, Edgar [/bib_ref] for maximum likelihood inference by PhyML v.3.3.20180214 [bib_ref] A simple, fast, and accurate algorithm to estimate large phylogenies by maximum..., Guindon [/bib_ref]. The proportion of invariable sites and the gamma distribution parameter were estimated from the data set, and the number of substitute rate categories was set to four. The bootstrap supports were estimated based on 1000 replicates.
## Analysis of the type vi secretion system genes
To identify the T6SS-associated genes, C58 [bib_ref] Agrobacterium tumefaciens deploys a superfamily of type VI secretion DNase effectors as..., Ma [/bib_ref] [bib_ref] VgrG C terminus confers the type VI effector transport specificity and is..., Bondage [/bib_ref] [bib_ref] Systematic dissection of the Agrobacterium type VI secretion system reveals machinery and..., Lin [/bib_ref] and other strains [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] [bib_ref] Redundancy and specificity of type VI secretion vgrG loci in antibacterial activity..., Santos [/bib_ref] that have been characterized experimentally were used as the references. Based on the known T6SS genes in these genomes, homologous genes in other genomes were identified based on the OrthoMCL result. To screen for novel T6SS effector, chaperone, and immunity genes, genes that are located near vgrG (i.e., three upstream and ten downstream) were examined manually by using the NCBI conserved domain database (CDD) [bib_ref] CDD: conserved domains and protein three-dimensional structure, Marchler-Bauer [/bib_ref] and the Phyre2 protein fold recognition server [bib_ref] The Phyre2 web portal for protein modeling, prediction and analysis, Kelley [/bib_ref] ; the e-value cutoff was set to 0.01. A few genes with a hit to the DUF4123 domain (i.e., a known T6SS chaperone) but have an e-value above the cutoff were manually added back to the list of T6SS-associated genes (e.g., CFBP5477_RS20350). To confirm the absence of specific T6SS genes, the genome sequences were used as the subjects and the protein sequences of known genes were used as the queries to run TBLASTN searches.
For classification of the vgrG homologs, we developed a domain-based scheme. The conserved N-terminal TIGR03361 domain was first identified by the NCBI CDD searches. A global alignment of all homologs was used to determine the exact boundaries of this domain. After this TIGR03361 domain was removed, the remaining C-terminal sequences were processed using MEME v5.1.1 [bib_ref] Suite: tools for motif discovery and searching, Bailey [/bib_ref] to identify conserved domains that meet these criteria: (1) present in at least two sequences, (2) with zero or one occurrence per sequence, (3) with a size between 30 and 300 a.a., and (4) with an e-value lower than 0.05. The results were manually curated to break down large domains that are composed of smaller domains. Pairwise BLASTP searches were conducted to verify that each domain is unique and no two domains have a BLASTP e-value of lower than 1e−05. For each domain, the consensus sequence was generated using Jalview v2.10.5 [bib_ref] Jalview version 2-a multiple sequence alignment editor and analysis workbench, Waterhouse [/bib_ref] and sequence conservation was visualized using WebLogo server v3 [bib_ref] WebLogo: a sequence logo generator, Crooks [/bib_ref]. For functional prediction, the consensus sequence of each domain was used to query against NCBI CDD and Phyre2. Additionally, one representative from each subtype of vgrG homologs was used for structure modeling using Phyre2 with the "normal" mode. The chain D of PA0091 VgrG1 (PDB identifier: 4MTK) was selected as the template. The predicted structures were visualized using PyMOL v1.2r3pre (Schrödinger, USA).
For the EI gene pairs identified, EI1 through EI11 were named based on the nomenclature proposed previously [bib_ref] Plantpathogenic Agrobacterium tumefaciens strains have diverse type VI effectorimmunity pairs and vary..., Wu [/bib_ref] , and novel pairs were named starting from EI12. When only the putative effector (E) or the putative immunity (I) genes were found, those genes were classified in the format of "E??" or "I??", respectively. For some of the EI pairs that were identified previously based on adjacency to T6SS genes but lacked high-confidence annotation (i.e., EI2, EI3, EI5, and EI8), we chose a more conservative approach and annotated those genes as hypothetical proteins.
## Analysis of the tumor-inducing plasmids and type iv secretion system genes
The list of 20 putative pTi sequences analyzed is provided in . These included all of the 15 complete sequences determined in this study and five representatives from GenBank that are important in Agrobacterium research. Our definition of putative pTi was based on the presence of the main T4SS genes (virB1-B11 and virD4) and at least one predicted T-DNA region. The pTi typing was performed based on k-mer profile clustering with a reference set of 143 oncogenic plasmids in Rhizobiaceae [bib_ref] Unexpected conservation and global transmission of agrobacterial virulence plasmids, Weisberg [/bib_ref] and a second set that contains > 4000 Rhizobiaceae plasmids [bib_ref] Diversification of plasmids in a genus of pathogenic and nitrogen-fixing bacteria, Weisberg [/bib_ref]. For T-DNA identification, putative T-DNA borders were identified based on the motif YGRCAGGATATATNNNNNKGTMAWN [bib_ref] Intragenic vectors for gene transfer without foreign DNA, Conner [/bib_ref]. Genes involved in opine metabolism [bib_ref] Opine biosynthesis and catabolism genes of Agrobacterium tumefaciens and Agrobacterium rhizogenes, Vladimirov [/bib_ref] and T4SS [bib_ref] The expanding bacterial type IV secretion lexicon, Bhatty [/bib_ref] were identified based on the annotation and homologous gene clustering results produced by OrthoMCL. Additionally, putative T4SS effectors were identified using the T4SEpre tool [bib_ref] Prediction of bacterial type IV secreted effectors by C-terminal features, Wang [/bib_ref] in EffectiveDB [bib_ref] EffectiveDB-updates and novel features for a better annotation of bacterial secreted proteins..., Eichinger [/bib_ref] with the minimal score set to 0.8. All protein sequences of pTi-encoded genes were used as the queries.
## Tumorigenesis assay
Tomato tumorigenesis assays [bib_ref] Expression and functional characterization of the Agrobacterium VirB2 amino acid substitution variants..., Wu [/bib_ref] were performed to evaluate the virulence of selected strains. The plants (cultivar Known-You 301) were maintained in growth chambers with a 16-/8-h light/dark regime and a constant temperature of 22°C. Inoculation was performed on 3-
week-old seedlings. Bacterial strains were transferred from stock to 5 mL 523 broth and cultured overnight at 28°C in a shaker incubator (250 rpm), then sub-cultured for 4 h prior to inoculation. Bacterial cells were washed and resuspended in 0.9% NaCl solution with a concentration of OD 600 0.2. The stem was punctured with a sterilized sewing needle, and 5 μL of bacterial suspension was added to the wounding site. The plants were collected 3 weeks after inoculation and 1-cm stem segments centered at the wounding site were cut for weighing.
## Supplementary information
The online version contains supplementary material available at https://doi. org/10.1186/s12915-021-01221-y.
Additional file 1: [fig_ref] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex [/fig_ref]. Gene content dissimilarity among the Agrobacterium genomes. (A) and (B): principal coordinate analysis with and without the outgroup A. larrymoorei, respectively. The % variance explained by each axis is provided in parentheses. (C) and (D): hierarchical clustering with and without the outgroup A. larrymoorei, respectively. . Global alignment of the linear chromosomes. Locations of T6SS-hcp operons and vgrG homologs are labeled. . Logo plots of the putative protein domains identified among vgrG homologs. For each domain, the length and the number of homologs with the domain is labeled. Domain 1 is the only domain with a corresponding database entry (TIGR03361). [fig_ref] Figure 4: Gene neighborhoods of vgrG homologs [/fig_ref]. Predicted structures of VgrG homologs. Regions are colored according to the scheme used in the domain analysis . The chain D of PA0091 VgrG1 (PDB identifier: 4MTK) from Pseudomonas aeruginosa was selected as the template. The C-terminal parts that could not be confidently inferred are omitted. In all cases, the coverage (i.e., percentage of the sequence included in the structure prediction) are at least 75%, the sequence identity to the template is at least 30% and the confidence score is 100%. [fig_ref] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes [/fig_ref]. Maximum likelihood phylogenies of vgrG-associated domains. (A) Domain 1 (TIGR03361; VI_Rhs_Vgr super family), (B) Domain 2 (unknown function), and (C) Domain 3 (unknown function). [fig_ref] Figure 6: Molecular phylogeny and global alignment of pTi [/fig_ref]. Principal coordinate analysis of gene content among the putative pTi analyzed. [fig_ref] Figure 7: Distribution of key vir genes among the putative pTi [/fig_ref]. Maximum likelihood phylogeny of pTi based on the concatenated alignment of shared single-copy genes. (A) All of the 20 pTi sequences analyzed; 21 core genes and 8534 aligned amino acid sites. (B) Excluding the two novel pTi; 40 core genes and 15,473 aligned amino acid sites, all branches received > 80% bootstrap support. . Tomato tumor assay of strains 12D1, CFBP4996, and CFBP5473. Mock was inoculated with sterilized water as a negative control and strain C58 was included as a positive control. Strain 12D1 harbors a plasmid with opine transporter and catabolism genes but lacks vir regulon genes and identifiable T-DNA. CFBP4996 and CFBP5473 harbor novel types of putative tumor-inducing plasmids (pTi). (A) Tomato stems at three weeks after inoculation. Scale bar: 0.25 cm. (B) Weight distribution of five biological replicates (1-cm segments of the stem centered at the inoculation site). The letters indicate ANOVA results. . Gene organization of the vir regulons on pTi. Syntenic regions are indicated by grey blocks. The virulence (vir) genes are highlighted in red, the conjugation (tra) genes are highlighted in yellow, and other genes are plotted in white.
Additional file 2: Dataset S1. List of vgrG-associated genes. Information including genomic location, RefSeq annotation, and domain prediction are included. Dataset S2. Locus tags of the vir regulon genes on pTi. The virA/J of 1D1609 are located on another plasmid and are highlighted by "*". The virB7 of pTiChry5 and pTiEU6 are unannotated in the GenBank RefSeq records so no locus tag is available but the gene presence was confirmed by BLASTN searches.
Kamoun provided helpful comments that improved the writing of this manuscript. The bacterial strains were imported under the permits 103-B-003 and 104-B-002 issued by the Council of Agriculture of Taiwan. The Sanger sequencing service and the Illumina sequencing library preparation service were provided by the Genomic Technology Core (Institute of Plant and Microbial Biology, Academia Sinica). The Illumina MiSeq sequencing service was provided by the Genomics Core (Institute of Molecular Biology, Academia Sinica). The PacBio sequencing and data processing service was provided by Genomics BioSci & Tech. Co. Ltd. (New Taipei City, Taiwan). The Institute of Plant and Microbial Biology (Academia Sinica) and the Department of Botany and Plant Pathology (Oregon State University) provided computing resources.
# Authors' contributions
[fig] Figure 1: Relationships among representatives of the Agrobacterium tumefaciens species complex. The sister species Agrobacterium larrymoorei (A. l.) is included as the outgroup. A Maximum likelihood phylogeny based on a concatenated alignment of 2093 single-copy genes shared by all 35 genomes (635,594 aligned amino acid sites). [/fig]
[fig] Figure 4: Gene neighborhoods of vgrG homologs. The grouping and labeling of vgrG homologs are based on the convention used in [/fig]
[fig] Figure 5: Phylogenetic distribution of vgrG homologs and T6SS effector/immunity genes. The species tree is based onFig. 1. Gene presence is illustrated with colored cells in the heatmap, gene copy numbers are labeled when applicable [/fig]
[fig] Figure 6: Molecular phylogeny and global alignment of pTi. The maximum likelihood phylogeny was inferred based on the concatenated alignment of 21 core genes and 8534 aligned amino acid sites (Supplementary [/fig]
[fig] Figure 7: Distribution of key vir genes among the putative pTi. The pTi sequences derived from this study are highlighted in bold. For those with relevant information available, the genomospecies assignments are indicated in square brackets and the opine types are indicated in parentheses. [/fig]
[fig] Funding: Research in the Chang lab was supported by the National Institute of Food and Agriculture, US Department of Agriculture awards 2014-51181-22384 and 2020-51181-32154. Research in the Lai lab was supported by Academia Sinica and the Ministry of Science and Technology of Taiwan (MOST 104-2311-B-001-025-MY3 and 107-2311-B-001-019-MY3). Research in the Kuo lab was supported by Academia Sinica and the Ministry of Science and Technology of Taiwan (MOST 109-2628-B-001-012 and 110-2628-B-001-020). The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication. Availability of data and materials The 14 new genome sequences are available in NCBI under BioProject accessions PRJNA534385-PRJNA534397 and PRJNA534399. [/fig]
[table] Table 1: List of the genome sequences used in this study. These include 14 new genomes derived from this study and 21 additional representatives from GenBank. Two Agrobacterium larrymoorei strains are included as the outgroup. Species name abbreviations: At, Agrobacterium tumefaciens; Al, Agrobacterium larrymoorei [/table]
|
Elder abuse in Europe’s “most elderly” city: an update of the phenomenon based on the cases reported to the Penal Court of Genoa from 2015 to 2019 and literature review
Background Elder abuse is currently a worldwide problem. The literature reports that one elderly person out of six is a potential victim. Aims To analyse cases reported to the judicial authorities in the territory of Genoa in the period 2010-2019, to investigate the features of elder abuse, to assess the trend of this phenomenon and to propose preventive strategies. Methods We analysed the data on reports of abuse passed by the Court of Genoa in the period 2015-2019 concerning physical and mental maltreatment, abandonment and financial exploitation of elderly subjects. These data were compared with those recorded in the previous 5-year period and in the literature.ResultsIn the period 2015-2019, 156 cases of elder abuse were identified (versus 63 in the previous period): 18 cases of domestic violence, 5 cases of abuse of the means of correction, 18 cases of caregiver neglect, 76 cases of physical injury and 39 cases of financial exploitation. Discussion Abuse was seen to be perpetrated most frequently in the domestic setting and by the victims' relatives. The main risk factors were female gender and the victim's dependence on others, the maltreating subject's mental illness and substance abuse. Conclusions We documented a progressive increase in the number of abuses reported to the judicial authority; this reflects greater awareness of the problem. However, our figures remained well below the incidence estimated in the literature. It is necessary to train healthcare personnel to identify and manage cases of suspected abuse, and to provide adequate support in situations at risk.Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
# Introduction
The phenomenon of elder abuse is a highly topical and significant issue. Indeed, it is estimated that, worldwide, one elderly person out of six is a victim of some type of abuse (about 141 million victims per year).
The WHO defines elder abuse as "a single or repeated act, or the lack of appropriate action, occurring within any relationship where there is an expectation of trust which causes harm or distress to an older person". The phenomenon is extremely serious in view of its possible impact on the life expectancy of elderly subjects; indeed, it gives rise to situations of mental and social distress and increases rates of mortality, hospitalisation and access to emergency departments.
Given that the world's population is steadily ageing, the phenomenon of elder abuse is destined to become increasingly common. Data on World Population Ageing indicate that the number of persons aged more than 65 years will increase from 703 million in 2019 to 1.5 billion in 2050. Moreover, it is estimated that the number of persons aged over 80 years will increase even more markedly, reaching 426 million in 2050. This latter age-group is at greater risk of frailty, isolation and impaired personal autonomy-factors associated with the risk of abuse. In Italy, according to the National Statistics Institute (ISTAT), by 2019 life expectancy at birth had risen to 81 years for men and 85.3 years for women. With regard to the metropolitan area of Genoa, one of Europe's "most elderly" cities and the focus of our investigation, elderly subjects (over 65 years of age) account for almost 29% of the total population of 841,180. Given the structure of the population and the high index of dependence (47.3%) of the elderly, the territory under examination is one in which the risk of abuse is high.
With regard to the phenomenon of elder abuse, we can distinguish abuse in domestic/community settingsand abuse in institutional settings (care facilities, hospitals).
Concerning abuse in the community setting, a recent meta-analysis reported that the phenomenon involved 15.7% of subjects aged over 65 years (11.6% psychological abuse, 6.8% financial abuse, 4.2% abandonment and neglect, 2.6% physical abuse, 0.9% sexual abuse).
With regard to abuse in institutional settings, the European Journal of Public Healthreported that almost two out of three (64.2%) staff members had reported some form of maltreatment of elderly patients, mainly physical and psychological. Moreover, a survey conducted among the staff of residences for the elderly in the United States in 1993 revealed that about 40% had perpetrated psychological abuse, while 10% had committed acts of physical violence. By contrast, a 2018 Chinese study of elderly people's residences reported a much lower incidence of the phenomenon: of the 681 elderly subjects recruited in two different facilities, 11.48% claimed to have suffered psychological abuse and 8.24% physical abuse.
In addition, a US study involving 2,713 elderly persons estimated the incidence of elder abuse in community settings to be about 8.8%; the risk factors identified were: female gender, advanced age (> 85 years) and the presence of multiple physical pathologies. The type of abuse most frequently cited was psychological (4.8%), followed by financial (2.9%), caregiver neglect (1.1%), physical (0.5%) and sexual abuse (0.1%).
The term "physical abuse" refers to the use of force that may result in injury, pain or permanent damage; these actions include violent behaviours, corporal punishment and the improper administration of pharmaceutical drugs.
Psychological maltreatment consists of causing anxiety and/or distress through the use of verbal or non-verbal language and includes threats, intimidation, molestation, and forced isolation of victims from their loved ones. The prolonged state of anguish that may ensue can result in changes of mood and even in full-blown depression.
Elderly people, who are often frail and dependent on others, are also vulnerable to theft, fraud or other types of economic exploitation. Financial abuse is defined as the illegal use of an elderly subject's funds, goods or property, the unauthorised cashing of cheques and falsification of the person's signature.
Another aspect of elder abuse is that of neglect. For instance, a caregiver may fail to provide for the elderly person's primary necessities, in terms not only of hygiene, feeding and therapy but also of psychological and social needs. This may be due to the caregiver's lack of professional training or to wilful neglect on the part of a caregiver who is only interested in being paid and/or obtaining free accommodation.
Risk factors for the maltreatment of elderly persons include: female sex, advanced age, cohabitation with the abuser, social isolation and a poor network of support; dementia and other neurological disorders also facilitate abuse, in that caring for such patients places a burden on the caregiver, both physically and, especially, psychologically.
For what concerns the perpetrators of abuse, they are most often members of the elderly person's own family, who are compelled to act as caregivers; this is mainly the case in community settings. In institutional settings, it is the healthcare workers themselves who act as caregivers, and who are therefore at risk of enacting abusive behaviours.
The factors that place caregivers at risk of becoming abusers are substance abuse (alcohol or drugs), psychiatric disorders, physical problems, social isolation, a perception of being overburdened with responsibility, and financial or employment problems that may make the subject dependent on the victim.
In Italy, the family is the mainstay of the care of elderly persons. Consequently, the phenomenon of "informal caregiving" is common; indeed, the elderly are frequently looked after by family members, who often lack the skills needed to carry out this task. Moreover, the commitment undertaken requires the caregiver to live in the same house as the elderly person and to make considerable sacrifices. Indeed, the burden of assistance is added to the ordinary duties of work and family; this may result in economic and social problems, while the elderly person's continual requests may give rise to stress, anxiety and depression.
Although the case-records reveal that elder abuse is a serious problem and that its dimensions are inevitably destined to increase, the phenomenon remains underestimated. Indeed, it has been claimed that only 1 in every 24 cases is reported to the judicial authorities. Although the institutions have devoted greater attention to elder abuse over the last decade (owing to media coverage of residential facilities for the elderly), the phenomenon has been only scantly investigated; this hinders the evaluation of its true magnitude and of the adoption of possible preventive measures.
In Italy, there is still no specific law that safeguards elderly persons. Indeed, under Italian law, the elderly subject is equated with any subject who is in a state of need, poverty or abandonment. In the Italian Penal Code, the articles that refer more specifically to violence against elderly persons are art. 570 (Violation of the obligations of family care), art. 571 (Abuse of the means of correction or discipline), art. 572 (Maltreatment of family members and cohabiting persons), art. 582 (Personal injuries), art. 591 (Abandonment of minors or incapable persons) and art. 643 (Circumvention of incapable persons). Owing to the intrinsic fragility of elderly persons, these subjects are vulnerable to the offences that we considered, the consequences of which, in prognostic terms, constitute a serious social and economic problem for society.
# Materials and methods
The aim of our study was to analyse cases of elder abuse covered by the above-mentioned articles of the Penal Code that came to the attention of the judicial authority in the metropolitan area of Genoa between 2010 and 2019, in order to examine the characteristics of this abuse and propose preventive strategies.
The data from the period 2015-2019 were obtained with the help of the Public Prosecutor's Office. Subsequently, these data were compared both with those recorded in a previous study conducted by the Department of Legal Medicine of the University of Genoa, which analysed reports and sentences registered in the previous 5-year period (2010-2014) , and with the data reported in the literature. In this way, we were able to depict the present state of the phenomenon and to trace its evolution over the last 10 years.
Examination of the contents of the sentences handed down enabled us to pick out several features of elder abuse in the territory considered: profiles of both abusers and victims (sex, age, psychological profile and possible pathologies of both), the type of relationship between the two, the settings in which the abuse took place and, finally, the verdicts pronounced (conviction or acquittal).
# Results
The following data emerged from the reports.
A total of 979 offences were reported in the period 2015-2019, as against 300 in the period 2010-2014.
With regard to the abuse of the means of correction or discipline, 7 reports were found; these involved offences of a strictly physical nature (beating, forced restraint, injury) that left visible signs on the victim's body; if such incidents are promptly reported, evidence of the offence can easily be gathered.
A large portion of the offences in our case-records involved maltreatment in the family setting, which accounted for 165 reports. Art. 572 safeguards the person's physical safety, mental salubrity and dignity, and refers to unlawful conduct that is reiterated over time and enacted with the intention of engendering a state of permanent oppression.
Of the offences considered, those involving personal injuries were the most common (504 reports). The cases examined referred to situations in which there was a considerable difference in age between the victim and the abuser, who exploited the elderly person's frailty.
The abandonment of an incapable elderly person (38 reports) was seen to stem chiefly from disputes among the children of an infirm elderly subject, each of whom may, for various reasons, refuse to take on the responsibility of caring for a parent who is no longer self-sufficient.
Finally, financial exploitation (240 reports) proved to be the second most frequent type of abuse, after physical and psychological maltreatment. Solitude, isolation and the presence of organic diseases that impair the individual's critical faculties and ability to distinguish between the good and bad intentions of others are the main risk factors.
It is very important to underline that there is a great difference between notifications of crimes and trial verdicts; indeed, while in the Italian legal system it is very simple to file a notification against someone, considerable evidence is needed to bring the person to trial. For this reason, there are many crime reports but few of them are judged worthy of prosecution. Analysis of the judicial sentences enabled us to construct profiles of both abusers and abused elderly persons (sex, age, physical and/or mental disorders) and to identify the causes and settings of the abuses committed. We analysed a total of 4500 sentences, only 156 of which involved the abuse of an elderly person (3.5% of the total, versus 2.1% in the previous 5-year period).
The following figures emerged . For what concerns art. 571 (Abuse of the means of correction or discipline) the 5 cases examined occurred in residential facilities for the elderly, the maltreating subjects being nurses or care workers. The unlawful behaviour was seen to have been enacted during night shifts and involved the imposition of uncomfortable positions for long periods, forced immobilisation, and abuse of the means of constraint. Only in one case, however, was the perpetrator convicted; in the other cases, working conditions were acknowledged to be arduous, with few staff members having to deal with large numbers of patients, and constraint devices being applied for long periods to control bouts of psychomotor agitation or delirium. What emerges here is that lack of time and staff shortages exacerbate care-related stress and reduce the quality of the care provided.
Analysis of the sentences regarding art. 572 (Maltreatment of family members and cohabiting persons) turned up 18 cases of abuse perpetrated in the victim's own home, which was shared by the abuser. In 84% of cases, the perpetrator was a son/daughter of the victim; most perpetrators (63%) were affected by problems of alcoholism, substance abuse or psychiatric disorders; moreover, 18 of the 21 maltreating subjects (86%) were males, and 15 of the 21 (71%) were aged over 40 years.
In the case of informal caregivers, a lack of training and an overload of responsibility may lead to conflict, which may then trigger violent behaviour (damage to property) and physical or verbal aggression (insults, humiliation, death threats). Many authors have identified the following risk factors of abuse: parent-child relationship, male sex of the abuser, age over 40 years, substance abuse, alcoholism, psychiatric disorders, financial dependence on the victim, and forced cohabitation between the victim and the abuser.
Victims' children also proved to be the main offenders in cases involving art 582 (Personal injury), with the home again being the most frequent setting of maltreatment. Moreover, the crime of personal injury is often associated with that of maltreatment in the family (art. 572). Indeed, situations of ongoing daily conflict may explode in physical violence that causes fractures, bruises or lacerations, with the result that treatment in the Emergency Department is required and the incident is reported to the judicial authority.
Three cases of neglect came to light. In these cases, the offender was the victim's caregiver, a subject who is specifically responsible for the care and supervision of the elderly person, and the setting of the maltreatment was the home of an elderly person, who was not self-sufficient. The behaviour took the form of dereliction of duty: long absences from the home, neglect and failure to maintain proper hygiene.
Financial abuse figured in 39 sentences. The victims suffered from mental deficiency (Parkinson's disease, dementia, depression) and had a poor social network. Age proved to be a major risk factor; of the 43 victims, 26 (60%) were over 80 years of age, while were aged between 75 and 84 years. The oldest old (> 85 years) proved to be at the greatest risk of financial abuse, often being isolated and dependent on others. The offenders did not have a daily relationship with their victims, but gained their trust to defraud them; this occurred in 25 cases involving strangers, 7 cases involving caregivers, and 3 cases involving the staff of residential facilities.
# Discussion and conclusions
Having analysed the results of the present study, and after comparing these with the data from the previous 5-year period [25] and those reported in the literature, we were able to draw several conclusions regarding the trend in the phenomenon of elder abuse over the past decade.
With regard to the sex of the victims of abuse, the literature indicates that women are at greater risk. Similarly, in our study, of the total of 249 elderly subjects who had suffered maltreatment, 134 were women and 115 were men.Moreover, women are at greater risk of reiterated abuse and physical and mental oppression.
Regarding the sex of abusers, there is no complete consensus in the literature. However, our study supported the conviction that a majority of cases of abuse are perpetrated by men. Indeed, on a total of 276 persons charged with abuse, 178 were men and 98 women; maltreatment perpetrated by males tended to be physical and violent (personal injury, abuse in the family), while females were more often involved in financial abuse.
Although physical and psychological abuse was seen to be perpetrated most frequently by a single victims' relative, regarding the extra-family context our study has proved that abuses (especially financial abuse) were mainly perpetrated by a group of two or three people. In fact, acting as real criminal organizations, these strangers were able to gain the confidence of the elderly (pretending to be lawyers, bankers or investors) and to obtain money, cheques, goods or property illegally.
The most common setting of abuse was seen to be the elderly person's own home; about 47% of our cases took place in the domestic environment. This can be explained by the fact that in Italy, unlike the US or Northern European countries (where many of the other studies have been conducted), infirm elderly people tend to be looked after at home, and only exceptionally in residential facilities. In our study, only 5 cases (8%) of abuse took place in institutional settings in the period 2010-2014, and only 14 (9%) in the period 2015-2019. Although estimates of the incidence of abuse in residential facilities in Italy are not yet available, such institutions seem to be safer than the domestic environment, not least on account of the adoption of modern systems of staff surveillance.
For what concerns the trend in the phenomenon of elder abuse over the decade considered, it must first be observed that the total number of cases notified differed markedly between the two 5-year periods; in the period 2015-2019, we recorded 979 reports and 156 sentences that met the criteria for inclusion in our analysis, as against 300 reports and 63 sentences in the previous 5-year period. This increase may have been due to the greater awareness of the problem on the part of the public and the judiciary, resulting in the emergence of cases that, in the past, would not have come to light.
The present study confirmed that the most frequent perpetrators of abuse were the children of the victims, followed by spouses and caregivers; in the previous study, however, caregivers occupied he second position. This finding could suggest that owing to greater public awareness of the phenomenon, the families of elderly persons now exercise greater caution in choosing caregivers.
Comparison of the sentences handed down by the Court of Genoa during the two 5-year periods also yielded interesting findings. The data that emerged from the two studies revealed that convictions had increased; during the second period, 51% of those charged with abuse were convicted, as against 35% in the first period. It can, therefore, be claimed that, during the period 2015-2019, more of the cases reported to the judiciary proved to be genuine cases of elder abuse, and that they were dealt with more strictly.
Given the progressive ageing of the global population, there is a need to increase our knowledge of the characteristics of maltreatment and its related mechanisms, to implement adequate prevention and to ensure prompt identification and effective management of the phenomenon. Doctors and the staff of emergency departments, hospitals and residential facilities play a fundamental role in the screening and identification of cases of abuse; this implies the need to implement screening protocols that are agreed upon by the scientific community.
Although the Comprehensive Geriatric Assessment (CGA) constitutes the "gold standard" for the thorough evaluation of the health and well-being of elderly subjects, it cannot be carried out in a short time (e.g. during visits by family doctors or non-geriatric specialists or in the emergency department). However, questionnaires like the BASE (Brief Abuse Screen for the Elderly) or the CASE (Caregiver Abuse Screen) can investigate suspected abuse during brief clinical examinations; these questionnaires differ in terms of their compilation time, ease of use and time required for personnel training.
The investigation of suspected abuse should include the history of the alleged maltreatment, the elderly person's recent and remote medical history, his/her dependence on others, assessment of the subject's functional capacity, identification of the use of physical or pharmacological means of constraint, psychiatric and cognitive evaluation, complete objective examination, and generic and toxicological laboratory tests. Caution must always be exercised to avoid confusing age-related physiological modifications with signs of abuse; the five main pathological conditions that may cause confusion are skin lesions, anomalous bleeding, fractures, malnutrition and anogenital lesions.
Interactions with healthcare personnel offer opportunities to recognise signs of abuse and enable suspected cases to be reported to the judicial authorities. Indeed, in the US, the percentage of accesses to the emergency department due to cases of ascertained elder abuse remains very low compared to the estimated prevalence. This reveals the need to rectify the shortcomings in the identification and reporting of cases of abuse, which could be done by introducing training courses for healthcare personnel and implementing clear and appropriate screening protocols.
In particular, health-care professionals should receive adequate clinical-forensic training in order not only to recognize cases of abuse but also to gather evidence that may be useful in criminal proceedings. If possible, after collecting the victim's details and story, consent shall be required for inspection, collection and correct storage of biological material for judicial purposes. After a general examination, the lesions on the victim's body must be located, accurately described and photographed, in order to ensure valuable evidence.
The ultimate objective, in fact, should be to improve knowledge of the phenomenon of elder abuse and to adopt a multidisciplinary approach to its management.
A further useful measure could be to introduce "emergency refuges", facilities where the elderly person can be temporarily assisted until such time as the problem can be properly dealt with. Moreover, systems of psychological and financial support should be made available to families who care for infirm elderly relatives.
With regard to cases of abuse by the staff of residential facilities, the causes appear to lie in the inadequate training of healthcare personnel and work-related stress. This implies that adequate training should be provided, workloads should be reduced and relationships between staff and inmates should be improved.
Elder abuse is an extremely complex and heterogeneous problem. Indeed, global prevention of the phenomenon, prompt identification of cases of maltreatment, the safeguard of victims and recognition of the difficulties facing caregivers are delicate issues that require adequate attention. Despite growing awareness of the phenomenon and the proposals mentioned above, elder abuse remains largely unrecognised and infrequently reported to the authorities. |
Identifying credible attribution sources for cigarette health warning labels in China: results from a cross-sectional survey of Chinese adults
GENERAL COMMENTSThis paper is a revision of a previously reviewed submission concerning ratings of credibility and perceived effectiveness of warning labels for cigarettes in China. A large sample of both current smokers and non-smokers were shown the same graphic image of a diseased lung on a mock cigarette package with text stating that smoking causes lung cancer. The packs were shown with 4 different variations that attributed the claim to different sources. Participants were asked simply to select the version that appeared most credible, most effective at making people quit and most effective at preventing young people from starting to smoke. While the revision has made the paper's findings clearer, there are still aspects that could be improved if the results are to be helpful for future efforts to develop warnings for China.
The WHO was seen as the most credible for the claim, but there was no difference between it and the Chinese CDC as most effective at getting smokers to quit or preventing young people from starting. The authors suggest in the abstract that the WHO being seen as more credible is important in this context. But I think that this is a misunderstanding of the question that was asked of participants. The credibility of the source refers to whether the claim is valid, and it seems that respondents tended to think that the WHO was a more credible source for the claim than the Chinese CDC. Whether the claim is motivating is another question, and unfortunately the study only provides tangential evidence about this. Asking people whether the message would encourage others to act is not the same as asking them whether it would motivate them to do that.
So, summarizing the findings as suggesting that they provide evidence for effectiveness of the messages is going a bit beyond what one can say about them. This is even more so for the item about preventing youth from smoking. Both of the effectiveness outcomes rely on what participants think others will do and while this is often used as a measure of message effect, it is not as valuable as knowing whether it might motivate smokers to consider quitting themselves. I think this needs to be spelled out more clearly in the Discussion and abstract.
The major takeaway from this study seems to be that warnings attributed to the Chinese tobacco regulatory agency or to the first lady are not as credible or potentially motivating as messages coming from the Chinese CDC or WHO. Whether warnings attributed to CDC or WHO would work at all remains to be seen in further testing.
I also find the use of the term relative risk to be confusing. There is no evidence of risk in these data and describing findings in those terms only seems to add confusion and potential for misunderstanding.
## Reviewer
Fang, Jennifer Simon Fraser University REVIEW RETURNED
## 22-mar-2022
## General comments
This is an interesting study with useful findings that can easily be translated into practice in China, which is yet to introduce HWL. The attribution sources tested represent an appropriate range of options. This is not critical, but I wondered why authors chose to select China CDC, rather than an equivalent of a Ministry of Health? It would have been useful to carry out the same study on rural populations, but this is appropriately addressed as one of the limitations, and understandable given the logistics and this particular exercise being nested within a bigger study.
A handful of more detailed suggestions below: (There are three different page numberings on the manuscript, so please note I used page numbers at the top left/right corners.) Page 6, line 31: China does not have a federal government. Please remove "federal" Page 6, lines 37-39: STMA is a government regulatory entity that administers the tobacco monopoly and oversees CNTC (production and import quotas, pricing etc). It is inaccurate to say they "enforce tobacco control regulation" as enforcement is dependent on the specific law being violated (ex. local municipal indoor smoking ban is often enforced by a local entity, whereas violations of the national Advertisement law is by the State Business Administration Authority). As a government entity, however, STMA would have a say at the policy table.
It may also be useful to clarify here that STMA is the regulatory of the tobacco industry (part of the government) and CNTC is the business arm (state-owned enterprise). The two are the same entity (and not just "generally considered" to be one)they share the same leadership, office and website. Page 8, line 17: remove "federal". "Central" would be more accurate Page 22, Thank you for raising this important point. We changed the language in the Abstract to better align with the outcomes measured. The Abstract conclusion now states:
"Conclusion Results suggest the unique role of health organizations in conveying smokingrelated messages that appear credible and effective at motivating others to quit smoking or never start smoking in China. Findings can inform global recommendations regarding HWL attribution sources."
In addition, we made edits throughout the Discussion (in red text in the marked copy) to indicate that the effectiveness measures were based on questions that asked participants to think about what others would do (e.g., quit smoking, prevent smoking) and not what the participant themselves would do.
We also added a Limitation statement that more directly acknowledges the limits of the effectiveness measure. The statement reads:
"Fifth, we did not measure other factors which may influence perceived credibility and our outcomes of interest, such as people's trust in government authorities or international health organizations. Finally, we did not assess how effective the messages would be on changing a participant's own behavior and only asked about effectiveness of changing the behavior of others. Future research should incorporate these additional credibility and effectiveness metrics into their study design and analysis."
We also acknowledge in the Conclusion of the Discussion that more research is needed to determine whether messages attributed to the WHO or China CDC would impact individual smoking behavior. The additional sentences in the Conclusion read:
"Our results suggest that HWL messages from an international public health authority such as WHO or a domestic health authority such as China CDC were viewed as more credible and motivating than messages attributed to the STMA and Peng Liyuan. It is possible that such message could influence smoking behavior, although future research is needed to fully understand the effectiveness of these warnings on individual behavior change (e.g., smoking cessation)."
1. The major takeaway from this study seems to be that warnings attributed to the Chinese tobacco regulatory agency or to the first lady are not as credible or potentially motivating as messages coming from the Chinese CDC or WHO. Whether warnings attributed to CDC or WHO would work at all remains to be seen in further testing.
In the Conclusion, we restate the importance of this finding and the limitation of the effectiveness measures that we used. The additional sentences in the Conclusion now read:
"Our results suggest that HWL messages from an international public health authority such as WHO or a domestic health authority such as China CDC were viewed as more credible and motivating than messages attributed to the STMA and Peng Liyuan. It is possible that such message could influence smoking behavior, although future research is needed to fully understand the effectiveness of these warnings on individual behavior change (e.g., smoking cessation)."
1. I also find the use of the term relative risk to be confusing. There is no evidence of risk in these data and describing findings in those terms only seems to add confusion and potential for misunderstanding.
We can appreciate that the terminology of 'crude relative risk' and 'adjusted relative risk ratio' might be unfamiliar or confusing to readers. Broadly, this is the exponentiated coefficient that is produced when using multinomial logistic regression. To help improve the reader's ability to interpret our results, we added the following description in the Methods Statistical Analysis section:
"We present results from the unadjusted and final adjusted models, where crude relative risk (RR) estimates and adjusted relative risk ratios (aRRRs) were considered significant at p<0.05. The crude RR estimate reflects the 'risk' or odds of the outcome falling into the comparison group (e.g., WHO) versus the reference group (e.g., China CDC) without controlling for any other predictive variables. The aRRR coefficient is a ratio of two probabilities and indicates how the 'risk' or odds of our outcome falling into the comparison group (e.g., WHO) compared to the 'risk' or odds of the outcome falling into the reference group (e.g., China CDC) changes with the predictive variable (e.g., age) included in the model. All analyses were completed using Stata version 17 (StataCorp, College Station, TX)."
## Reviewer 2:
1. This is an interesting study with useful findings that can easily be translated into practice in China, which is yet to introduce HWL. The attribution sources tested represent an appropriate range of options. This is not critical, but I wondered why authors chose to select China CDC, rather than an equivalent of a Ministry of Health?
Thank you for this important question. The Ministry of Health equivalent in China, known as the National Health Commission, was officially formed in 2018. At the time of our data collection, this organization was known as the National Health and Family Planning Commission and the level of engagement in tobacco control activities was not widely known. However, the China CDC was actively engaged with tobacco control efforts. We believed their voice might be credible and their name could more easily communicate to a lay person that they are a governmental body involved in preventing disease. We added additional detail to support the selection of China CDC to the Introduction. The description of the four sources now reads:
"This study examined the perceived credibility and perceived effectiveness at reducing tobacco use of four different Chinese HWL message sources: STMA (the domestic tobacco regulatory agency), the China Center for Disease Control (China CDC; a federal government health agency actively engaged in tobacco control and disease prevention activities at the time of data collection), the World Health Organization (WHO; a global health authority), and China's First Lady Peng Liyuan, who is a celebrity/public figure and has been the Chinese Association on Tobacco Control ambassador since 2009."
## It would have been useful to carry out the same study on rural populations, but this is appropriately addressed as one of the limitations, and understandable given the logistics and this particular exercise being nested within a bigger study.
We agree this is an important area of future study and appreciate your assessment that we have appropriately addressed this limitation.
## China does not have a federal government. please remove "federal"
We removed the word "federal" throughout the manuscript.
1. STMA is a government regulatory entity that administers the tobacco monopoly and oversees CNTC (production and import quotas, pricing etc). It is inaccurate to say they "enforce tobacco control regulation" as enforcement is dependent on the specific law being violated (ex. local municipal indoor smoking ban is often enforced by a local entity, whereas violations of the national Advertisement law is by the State Business Administration Authority). As a government entity, however, STMA would have a say at the policy table. It may also be useful to clarify here that STMA is the regulatory of the tobacco industry (part of the government) and CNTC is the business arm (state-owned enterprise). The two are the same entity (and not just "generally considered" to be one)they share the same leadership, office and website.
Thank you for these nuanced corrections. We revised this sentence to know read:
"In China, the tobacco industry is a part of the federal governmentcigarettes are manufactured, distributed and sold by the state-owned Chinese National Tobacco Company (CNTC), and the State Tobacco Monopoly Administration (STMA) is the regulatory arm that oversees CNTC activities and plays a role in policymaking.[6, 7]"
1. Page 8, line 17: remove "federal". "Central" would be more accurate
We removed the word "federal" throughout the manuscript.
## General comments
Thank you for making the changes. I would still prefer describing the statistics as the odds of rejecting the null or something like that rather than relative risk, but I leave it to you to make that call.
## Reviewer
## Fang, jennifer simon fraser university
## Review returned
06-Apr-2022
## General comments
A couple of minor comments: p. 11 of 53, CDC "was actively engaged in tobacco control and disease prevention efforts" is repeated twice in the first paragraph, on lines 15-16 and 23-24. Suggest only keeping one and removing the other. p. 11 of 53. I note the addition of an explanation of RR and aRRR and think this is great for clarity.
p. 15 pages 19-43. In this paragraph, authors discuss the credibility of STMA as a regulatory source to attribute health warnings to. This may be my bias as a tobacco control researcher, however, I have strong reservations about likening the STMA (part of the tobacco industry) to a government regulatory body like the FDA, and even referring to the STMA as a "regulatory agency". However, I also do accept that the nuances of the industry structure may not be obvious to a regular person. (This is just a comment and something for authors to consider.)
## Version 2 -author response
Reviewer 1 comments:
1. Thank you for making the changes. I would still prefer describing the statistics as the odds of rejecting the null or something like that rather than relative risk, but I leave it to you to make that call.
Thank you for this honest feedback. We plan to leave the results written as is; however, we hope that the additional description of relative risks in the Methods improved clarity and interpretability.
## Reviewer 2 comments:
1. p. 11 of 53, CDC "was actively engaged in tobacco control and disease prevention efforts" is repeated twice in the first paragraph, on lines 15-16 and 23-24. Suggest only keeping one and removing the other.
Thank you for pointing this out. The additional fragment has been deleted.
1. p. 11 of 53. I note the addition of an explanation of RR and aRRR and think this is great for clarity.
Thank you for this feedback. We are glad that the explanation improved clarity of the response. Thank you for this important point of feedback. We are now clear that the STMA and CNTC are one one-in-the-same entity. Throughout the text (in track changes) we refer to "STMA as the regulatory arm of the state tobacco company." Further we revised our Discussion of results regarding perceived credibility of messages from STMA. We now state:
"In contrast, agencies whose charge it is to regulate tobacco productswhether an independent governing body or integrated with the tobacco industry like STMAmay not be as clearly associated with health-related activities. For example, studies from the United States suggest that attributing warnings to the tobacco regulatory agency, the US Food and Drug Administration, does not increase their believability. In China, it is also possible that our study participants are aware of and view the STMA as the same organization as CNTC given the complex national system of tobacco production and regulation,[7] and thus, potentially rated the STMA as less credible given its tobacco industry ties.
[table] Table 1 ,: So, summarizing the findings as suggesting that they provide evidence for effectiveness of the messages is going a bit beyond what one can say about them. This is even more so for the item about preventing youth from smoking. Both of the effectiveness outcomes rely on what participants think others will do and while this is often used as a measure of message effect, it is not as valuable as knowing whether it might motivate smokers to consider quitting themselves. I think this needs to be spelled out more clearly in the Discussion and abstract. [/table]
|
On the Stability of One-Dimensional Wave Equation
We prove the generalized Hyers-Ulam stability of the one-dimensional wave equation, = 2 , in a class of twice continuously differentiable functions.
# Introduction
In 1940, Ulamgave a wide ranging talk before the mathematics club of the University of Wisconsin in which he discussed a number of important unsolved problems. Among those was the question concerning the stability of group homomorphisms:
Let 1 be a group and let 2 be a metric group with the metric (⋅, ⋅). Given > 0, does there exist a > 0 such that if a function ℎ : 1 → 2 satisfies the inequality (ℎ( ), ℎ( )ℎ( )) < , for all , ∈ 1 , then there exists a homomorphism : 1 → 2 with (ℎ( ), ( )) < , for all ∈ 1 ?
The case of approximately additive functions was solved by Hyers [bib_ref] On the stability of the linear functional equation, Hyers [/bib_ref] under the assumption that 1 and 2 are Banach spaces. Indeed, he proved that each solution of the inequality ‖ ( + ) − ( ) − ( )‖ ≤ , for all and , can be approximated by an exact solution, say an additive function. In this case, the Cauchy additive functional equation, ( + ) = ( ) + ( ), is said to have the Hyers-Ulam stability. Rassias [bib_ref] On the stability of the linear mapping in Banach spaces, Rassias [/bib_ref] attempted to weaken the condition for the bound of the norm of the Cauchy difference as follows:
[formula] ( + ) − ( ) − ( ) ≤ (‖ ‖ + )(1) [/formula]
and proved Hyers' theorem. That is, Rassias proved the generalized Hyers-Ulam stability (or Hyers-Ulam-Rassias stability) of the Cauchy additive functional equation. Since then, the stability of several functional equations has been extensively investigated [bib_ref] Hyers-Ulam stability of functional equations in several variables, Forti [/bib_ref] [bib_ref] A generalization of the Hyers-Ulam-Rassias stability of approximately additive mappings, Gȃvrutȃ [/bib_ref] [bib_ref] Approximate homomorphisms, Hyers [/bib_ref] [bib_ref] On the stability of functional equations and a problem of Ulam, Rassias [/bib_ref].
The terminologies, the generalized Hyers-Ulam stability, and the Hyers-Ulam stability can also be applied to the case of other functional equations, differential equations, and various integral equations.
Given a real number > 0, the partial differential equation
[formula] ( , ) − 2 ( , ) = 0 (2) [/formula]
is called the (one-dimensional) wave equation, where ( , ) and ( , ) denote the second time derivative and the second space derivative of ( , ), respectively.
Let
[formula] : R × R → [0,∞) be a function. If, for each twice continuously differentiable function : R × R → C satisfying ( , ) − 2 ( , ) ≤ ( , ) ( , ∈ R) ,(3) [/formula]
there exist a solution 0 : R × R → C of the (one-dimensional) wave equation (2) and a function Φ :
[formula] R×R → [0, ∞) such that ( , ) − 0 ( , ) ≤ Φ ( , ) ( , ∈ R) ,(4) [/formula]
where Φ( , ) is independent of ( , ) and 0 ( , ), then we say that the wave equation (2) has the generalized Hyers-Ulam stability (or the Hyers-Ulam-Rassias stability). In this paper, using an idea from [bib_ref] On the stability of Laplace's equation, Hegyi [/bib_ref] , we prove the generalized Hyers-Ulam stability of the (one-dimensional) wave equation (2).
## Generalized hyers-ulam stability
In the following theorem, using the d' Alembert method (method of characteristic coordinates), we prove the generalized Hyers-Ulam stability of the (one-dimensional) wave equation [bib_ref] On the stability of the linear functional equation, Hyers [/bib_ref].
[formula] Theorem 1. Let a function : R × R → [0, ∞) be given such that the double integral ∫ 0 ∫ 0 ( + ] 2 , − ] 2 ) ](5) [/formula]
exists for all , ∈ R. If a twice continuously differentiable function : R × R → C satisfies the inequality
[formula] ( , ) − 2 ( , ) ≤ ( , )(6) [/formula]
for all , ∈ R, then there exists a solution 0 : R × R → C of the wave equation (2) which satisfies
[formula] ( , ) − 0 ( , ) ≤ 1 4 2 ∫ − 0 ∫ + 0 ( + ] 2 , − ] 2 ) ](7) [/formula]
for all , ∈ R.
Proof. Let us define a function V :
[formula] R × R → C by V ( , ) := ( + 2 , − 2 ) .(8) [/formula]
for all , ∈ R. Hence, we have
[formula] ( , ) − 2 ( , ) = −4 2 V ( , ) ,(10) [/formula]
for any , ∈ R. Thus, it follows from inequality (6) that
[formula] V ( , ) ≤ 1 4 2 ( + 2 , − 2 ) ,(11) [/formula]
for any , ∈ R. Therefore, we get
[formula] − 1 4 2 ∫ 0 ∫ 0 ( + ] 2 , − ] 2 ) ] ≤ ∫ 0 ∫ 0 V ( , ]) ] ≤ 1 4 2 ∫ 0 ∫ 0 ( + ] 2 , − ] 2 ) ](12) [/formula]
or equivalently
[formula] |V ( , ) − V ( , 0) − V (0, ) + V (0, 0)| ≤ 1 4 2 ∫ 0 ∫ 0 ( + ] 2 , − ] 2 ) ] ,(13) [/formula]
for all , ∈ R.
On account of (8), we get
[formula] V ( , ) = ( + 2 , − 2 ) , V ( , 0) = ( 2 , 2 ) , V (0, ) = ( 2 , − 2 ) , V (0, 0) = (0, 0) .(14) [/formula]
Hence, it follows from (13) and the last equalities that
[formula] ( + 2 , − 2 ) − ( 2 , 2 ) − ( 2 , − 2 ) + (0, 0) ≤ 1 4 2 ∫ 0 ∫ 0 ( + ] 2 , − ] 2 ) ] ,(15) [/formula]
for all , ∈ R. If we set = + and = − in the last inequality, then we obtain
[formula] ( , ) − 0 ( , ) ≤ 1 4 2 ∫ − 0 ∫ + 0 ( + ] 2 , − ] 2 ) ] ,(16) [/formula]
for all , ∈ R, where we set 0 ( , ) := ( 2 + 2 , 2 + 2 )
[formula] + ( 2 − 2 , − 2 + 2 ) − (0, 0) .(17) [/formula]
for all , ∈ R. Hence, we know that
for any , ∈ R; that is, 0 ( , ) is a solution of the wave equation (2).
## Corollary 2.
Given a constant > 0, let a function : R × R → [0, ∞) be given as
[formula] ( , ) = − 2 − 2 2 .(20) [/formula]
If a twice continuously differentiable function : R × R → C satisfies inequality, for all , ∈ R, then there exists a solution 0 : R × R → C of the wave equation (
for all , ∈ R.
Proof. Since
[formula] ∫ 0 ∫ 0 ( + ] 2 , − ] 2 ) ] = ∫ 0 ∫ 0 − 2 /2−] 2 /2 ] = ∫ 0 −] 2 /2 ] ∫ 0 − 2 /2 = 2 4 ( 2 √ ∫ /√2 0 −] 2 ]) ( 2 √ ∫ /√2 0 − 2 ) = 2 erf ( √ 2 ) erf ( √ 2 ) < ∞,(22) [/formula]
for all , ∈ R, in view of Theorem 1, we conclude that the statement of this corollary is true. |
Post-traumatic stress disorder and post-traumatic stress symptoms following critical illness in medical intensive care unit patients: assessing the magnitude of the problem
Introduction Post-traumatic stress disorder (PTSD) is a potentially serious psychiatric disorder that has traditionally been associated with traumatic stressors such as participation in combat, violent assault, and survival of natural disasters. Recently, investigators have reported that the experience of critical illness can also lead to PTSD, although details of the association between critical illness and PTSD remain unclear.MethodsWe conducted keyword searches of MEDLINE and Psych Info and investigations of secondary references for all articles pertaining to PTSD in medical intensive care unit (ICU) survivors.ResultsFrom 78 screened papers, 16 studies (representing 15 cohorts) and approximately 920 medical ICU patients met inclusion criteria. A total of 10 investigations used brief PTSD screening tools exclusively as opposed to more comprehensive diagnostic methods. Reported PTSD prevalence rates varied from 5% to 63%, with the three highest prevalence estimates occurring in studies with fewer than 30 patients. Loss to followup rates ranged from 10% to 70%, with average loss to followup rates exceeding 30%.
# Introduction
Estimates of post-traumatic stress disorder (PTSD) prevalence in critically ill cohorts are reported to be as high as 63% [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] and exceed or rival those of traditionally 'high-risk' populations as well as populations with medical disorders such as cancer and myocardial infarction [bib_ref] Post-traumatic stress disorder in cancer: a review, Smith [/bib_ref] [bib_ref] Symptoms of posttraumatic stress disorder in patients who have had a myocardial..., Shemesh [/bib_ref] [fig_ref] Table 1: A comparison of PTSD prevalence rates across 'at-risk' adult populations Studies listed... [/fig_ref]. It may be that critical illness is uniquely stressful due to factors associated with the intensive care unit (ICU) experience such as awareness during painful procedures, a sense of helplessness, loss of control, and an imminent threat of death. Such experiences may be 'traumatic' as trauma is a generic term that can refer to experiences that are physical and/or psychological in nature. Alternatively, it may be that the limited research conducted to date has substantially overestimated the prevalence of PTSD after critical illness or that PTSD in ICU survivors is qualitatively different than that resulting from war, natural disasters, or other types of traumatic stressors. A comprehensive evaluation of this and other issues is timely and important as concern about PTSD among ICU survivors is growing and has led, in some cases, to changes in the delivery of care and in the management of patients in response to the perception that PTSD is a common outcome.
A number of recent reviews have looked at the association between medical illness and the development of psychiatric illness [bib_ref] Posttraumatic stress disorder following medical illness and treatment, Tedstone [/bib_ref] [bib_ref] Epidemiology and treatment of psychiatric conditions that develop after critical illness, Weinert [/bib_ref] [bib_ref] Chronic neurocognitive effects of critical illness, Hopkins [/bib_ref]. However, no review has focused exclusively and/ or comprehensively on PTSD following medically related critical illness. With this review, we sought to accomplish four goals: (a) to evaluate existing research pertaining to PTSD following medically related critical illness, with a primary focus on prevalence, (b) to provide a critical analysis of methodological characteristics of the studies under review, (c) to provide a summary of possible explanations for PTSD following critical illness, and (d) based upon an analysis of the strengths and weaknesses of existing investigations, to offer recommendations for future research. For a definition of PTSD, see .
# Materials and methods
## Study identification and selection
A literature search for all articles pertaining to critical illness and PTSD was conducted using both the Psych Info and US National Library of Medicine MEDLINE databases. Key words/ phrases used to search these databases included 'post-traumatic stress disorder' AND 'critical illness' (25 abstracts via MEDLINE and 5 via Psych Info) or 'post-traumatic stress disorder' AND 'intensive care' (81 abstracts via MEDLINE and 19 via Psych Info). Reference lists from identified articles were used to identify any additional studies.
## Study inclusion criteria and evaluation
For inclusion in this review, studies were required (a) to evaluate the association between medical ICU hospitalization and PTSD (either the diagnostic entity called PTSD or post-traumatic stress symptoms and (b) to employ qualitative and/or objective measures of PTSD or PTSS. Investigations published in a language other than English were excluded as were unpublished studies and abstracts. One of the authors (JCJ) reviewed all of the articles in question to ensure that they met the above criteria.
## Data extraction and analysis
The following aspects of each study were identified, abstracted, and analyzed: study population, study design, timing of evaluations, study aims, exclusion criteria, methods of assessing PTSD, and all relevant results compared across study populations, including follow-up rates. All individual articles were assigned a 'quality rating' according to the Oxford Centre for Evidence-Based Medicine guidelines for symptom prevalence studies . Ratings ranged from 1 to 3, with lower numbers indicating higher quality.
# Results
## Search for articles
A total of 78 non-overlapping potential abstracts were identified in the search of the databases and reference lists (the most recent search was performed in October 2006). Of these, 16 papers met inclusion criteria [fig_ref] Table 3: Studies that report the prevalence of PTSD in medical ICU patients [/fig_ref]. A number of studies consisting entirely of physical trauma and/or surgical ICU patients were identified and excluded from review due to the likelihood that the PTSD symptoms experienced by these patient populations could have been generated by either trauma-related injuries or surgical interventions. The authors recognize that trauma and surgical ICU patients may be similar in many respects to their medical ICU counterparts and, indeed, they may have overlapping experiences. Nevertheless, we chose to exclude such patients so as to focus as specifically as possible on the unique contributions of medically related critical illness to the development of PTSD. Similarly, a number of research investigations of medical ICU survivors assessing anxiety or memories of the ICU generically were identified and were also excluded as they did not include a specific focus on PTSD or PTSS. One investigation evaluated PTSD symptoms after critical illness but did not include data regarding prevalence rates and thus was excluded [bib_ref] Memory, delusions, and the development of acute posttraumatic stress disorder-related symptoms after..., Jones [/bib_ref].
## Methods of reviewed articles
## Subject characteristics
All investigations were conducted exclusively on adult critically ill patients. Studies focused on general medical ICU populations [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] [bib_ref] Rehabilitation after critical illness: a randomized, controlled trial, Jones [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Survival, morbidity, and quality of life after discharge from intensive care, Eddleston [/bib_ref] [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref] as well as on critically ill patients with specific medical conditions such as ARDS/acute lung injury and septic shock [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] [bib_ref] Linguistic analysis to asses medically related posttraumatic stress symptoms, Shaw [/bib_ref] [bib_ref] Intensive care unit drug use and subsequent quality of life in acute..., Nelson [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref]. Within individual studies, patients had significant variability with regard to key characteristics such as ICU length of stay, ventilation status and duration of mechanical ventilation, severity of illness, and the time to PTSD assessment. One investigation included patients with ICU lengths of stay from 11 to 99 days [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref]. Another study included both patients with and without mechanical ventilation as well as those with APACHE II (Acute Physiology and Chronic Health Evaluation II) scores ranging from 4 to 38, suggesting extreme differences in illness severity [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref]. In a third investigation, follow-up evaluations were conducted at intervals ranging from 1 to 13 years [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref].
## Study design
A total of six studies were prospective in nature; five of these were cohort studies [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] Survival, morbidity, and quality of life after discharge from intensive care, Eddleston [/bib_ref] [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref] and one was a randomized controlled trial [bib_ref] Rehabilitation after critical illness: a randomized, controlled trial, Jones [/bib_ref]. Six investigations employed a retrospective cohort design [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Social support during intensive care unit stay might reduce the risk for..., Deja [/bib_ref]. Four studies were cross-sectional [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Linguistic analysis to asses medically related posttraumatic stress symptoms, Shaw [/bib_ref] [bib_ref] Intensive care unit drug use and subsequent quality of life in acute..., Nelson [/bib_ref]. Sample sizes were universally small, and the number of patients participating in follow-up ranged from 20 [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Linguistic analysis to asses medically related posttraumatic stress symptoms, Shaw [/bib_ref] to 143 [bib_ref] Survival, morbidity, and quality of life after discharge from intensive care, Eddleston [/bib_ref] patients. Four studies evaluated individuals at multiple time points, and initial evaluations occurred within two months of hospital discharge and followup evaluations occurred at widely varying intervals of up to eight years [bib_ref] Rehabilitation after critical illness: a randomized, controlled trial, Jones [/bib_ref] [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref] [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref]. The remaining investigations evaluated patients at a single time point, ranging from 3 months to 13 years after hospital or ICU discharge [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Survival, morbidity, and quality of life after discharge from intensive care, Eddleston [/bib_ref] [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] [bib_ref] Linguistic analysis to asses medically related posttraumatic stress symptoms, Shaw [/bib_ref] [bib_ref] Intensive care unit drug use and subsequent quality of life in acute..., Nelson [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Social support during intensive care unit stay might reduce the risk for..., Deja [/bib_ref]. The percentage of patients lost to follow-up (for any reason) varied from 16% [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] to 70% [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] , and the average rate of loss to follow-up was 32.5%. Three samples consisted DSM-IV definition of post-traumatic stress disorder
Definition of post-traumatic stress disorder a A potentially debilitating psychiatric condition that develops as the result of being exposed to a traumatic occurrence 'in which a person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury, or a threat to the physical integrity of self or others' and which generates 'intense feelings of fear, helplessness, or horror' in those exposed to the trauma. This condition is characterized by a constellation of symptoms in three domains:
A. Symptoms of re-experiencing (for example, intrusive thoughts and upsetting recollections of the trauma, recurrent dreams or nightmares, and flashbacks).
B. Symptoms of avoidance and emotional numbing (for example, efforts to avoid conversations, places, and thoughts associated with the trauma; detachment from others; and a restricted range of affect).
C. Symptoms of increase arousal (for example, sleep disruption, hypervigilance, and exaggerated startle response).
These symptoms must meet two criteria to satisfy diagnostic criteria:
1. Symptoms must cause significant impairment in social, occupational, or other important functional domains.
2. Symptoms must be present for at least 1 month after exposure to the traumatic event or events. Deeper levels of sedation and neuromuscular blockade exposure associated with increased risk of PTSD a Quality of study methods was rated according to Oxford Centre for Evidence-Based Medicine guidelines and ranged from 1 to 3, with lower numbers indicating higher quality. Letters used to designate level 1 to 3 studies indicated gradations of quality ranging from 'a' (higher quality) to 'b' (lower quality). b Total number of patients who were actual study participants as opposed to those who were simply enrolled; percentage lost to follow-up refers to the percentage of patients who for any reason did not participate in the follow-up portion or portions of the study. A few studies did not include follow-up components, thus loss to follow-up rates are not applicable (N/A). c Fourteen patients in the 2001 study of Schelling et al. [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] had previously been in the 1999 investigation of Schelling et al. [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref]. d These investigations were conducted on the same population, and the follow-up evaluations in the 1999 study of Stoll et al. [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] occurred approximately 2 years after patients completed their participation in the 1998 study of Schelling et al. [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref]. of patients who were five or more years apart with regard to time from ICU or hospital discharge [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref].
## Exclusion criteria/identification of pre-existing psychiatric illness
Studies in which exclusion criteria were stated explicitly included prior psychiatric illness or neurologic trauma or disease [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Rehabilitation after critical illness: a randomized, controlled trial, Jones [/bib_ref] [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref]. Methods of identifying pre-existing psychiatric disorders varied widely across studies, and only five studies formally inquired about patients' pre-morbid psychiatric histories [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref]. One of these investigations included a single question about pre-morbid psychiatric history, and this regarded whether subjects had seen a mental health professional or general practitioner for psychiatric reasons prior to ICU hospitalization [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref].
## Methods of assessing ptsd
A total of nine investigations relied solely on standardized brief screening tools in their assessment of PTSD or PTSS, including the Post-Traumatic Stress Scale-10 for the ICU (PTSS-10), Impact of Events Scale (IES), IES Revised, Davidson Trauma Scale (DTS), Trauma Symptom Checklist-33, and the Experiences of Treatment in the Intensive Care-7 [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] Rehabilitation after critical illness: a randomized, controlled trial, Jones [/bib_ref] [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] [bib_ref] Linguistic analysis to asses medically related posttraumatic stress symptoms, Shaw [/bib_ref] [bib_ref] Social support during intensive care unit stay might reduce the risk for..., Deja [/bib_ref]. With the exception of two investigations, these tests were administered in person [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Social support during intensive care unit stay might reduce the risk for..., Deja [/bib_ref]. Diagnoses of PTSD were repeatedly made entirely on the basis of information derived from screening tools. For example, Cuthbertson and colleagues [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] reported that 14% of their subjects met full diagnostic criteria for PTSD, despite the fact that the DTS (used in their investigation) is not a diagnostic tool. Similarly, Schelling and colleagues [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] diagnosed nearly 30% of ARDS survivors with PTSD on the basis of a cutoff score as opposed to a formal clinical interview. Few studies attempted to identify or quantify the clinical significance of PTSD or to evaluate commonly studied outcomes in this regard (for example, increased health care use, increased marital or family conflict, substance abuse, and days away from work), although three investigations did focus on the association between PTSD and health-related quality of life [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref].
## A total of five investigations relied on structured clinical interviews such as the structured clinical interview for the dsm-iv (diagnostic and statistical manual of mental disorders,
Fourth Edition) (SCID) [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] , employing them after screening tools were suggestive of probable PTSD. Generally, the use of more comprehensive tools such as the SCID resulted in the identification of fewer cases. For example, in the study by Nickel and colleagues [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] , approximately half of the subjects identified as having PTSD via the PTSS-10 were false-positive according to the SCID.
## Primary findings
How prevalent is ICU-related PTSD? Prevalence rates ranged from 5% to 63% and showed little variance regardless of whether the outcome in question was PTSD or PTSS; the three highest rates (54%, 59%, and 63%) occurred in investigations that purported to diagnose PTSD [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref]. Importantly, these rates were reported in subpopulations (control groups) with sample sizes of between 11 and 27 patients and were higher than the rates reported in their entire populations. Prevalence rates varied depending on the time of assessment and were highest at the time of hospital discharge or shortly thereafter, decreasing over time. For example, Kapfhammer and colleagues [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] reported that 43.5% of study subjects had PTSD at hospital discharge whereas 23.9% suffered from PTSD an average of eight years later.
General medical ICU cohorts had both the lowest and highest rates of PTSD or PTSS compared with more specialized populations. In studies of general medical ICU patients, prevalence rates ranged from 5%to 63% [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] , and rates in specialized populations ranging from 18.5% [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] to 43% [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref]. In the three studies comparing patients from different treatment conditions [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] , marked differences in prevalence rates existed between 'treatment' and 'control' arms.
## Risk factors for ptsd
Risk factors were not studied systematically across studies, although a number of risk factors were identified [fig_ref] Table 4: PTSD risk factors reported in the ICU-and PTSD-related literature at largeKnown risk... [/fig_ref]. Two investigations reported that delusional memories (as opposed to factual ones) increased the risk of PTSD [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] [bib_ref] Rehabilitation after critical illness: a randomized, controlled trial, Jones [/bib_ref] , and another study supported a relationship between fewer factual memories and a greater likelihood of PTSD. Alternatively, three studies implicated factual memories in the development of PTSD [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] Sensitivity and specificity of a screening test to document traumatic experiences and..., Stoll [/bib_ref] [bib_ref] Health-related quality of life and posttraumatic stress disorder in survivors of the..., Schelling [/bib_ref] , reporting an association between the number of traumatic memories and higher scores on PTSD screening tools. One study reported that greater recall of ICU-related experiences was associated with more intrusive symptoms [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref]. One study reported an association between the presence of anxiety in the ICU and symptoms of PTSD [bib_ref] Social support during intensive care unit stay might reduce the risk for..., Deja [/bib_ref]. Hospital-or treatment-related variables associated with PTSD or PTSS were associated with increased length of stay and/or duration of mechanical ventilation [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref] [bib_ref] Posttraumatic stress disorder and health-related quality of life in long-term survivors of..., Kapfhammer [/bib_ref] as well as greater levels of sedation and/or neuromuscular blockade [bib_ref] The long-term psychological effects of daily sedative interruption on critically ill patients, Kress [/bib_ref] [bib_ref] Intensive care unit drug use and subsequent quality of life in acute..., Nelson [/bib_ref]. Hydrocortisone treatment was associated with a decreased risk of PTSD in two investigations [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref] [bib_ref] The effect of stress doses of hydrocortisone during septic shock on posttraumatic..., Schelling [/bib_ref].
Demographic and historical variables associated with an increased risk of PTSD or PTSS included younger age [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Predictors of emotional outcomes of intensive care, Rattray [/bib_ref] , a prior mental health history [bib_ref] Post-traumatic stress disorder after critical illness requiring general intensive care, Cuthbertson [/bib_ref] [bib_ref] The occurrence of posttraumatic stress disorder in patients following intensive care treatment:..., Nickel [/bib_ref] , and female gender [bib_ref] Psychological problems following ICU treatment, Scragg [/bib_ref] [bib_ref] Survival, morbidity, and quality of life after discharge from intensive care, Eddleston [/bib_ref]. A greater degree of perceived social support was reported to be protective against the development of PTSS [bib_ref] Social support during intensive care unit stay might reduce the risk for..., Deja [/bib_ref].
# Discussion
Challenges to studying PTSD As others have observed, PTSD, as concurrently conceptualized by the DSM-IV and the psychiatric community, is a complex condition that presents unique diagnostic challenges for clinical researchers [bib_ref] Posttraumatic stress disorder part II: development of the construct within the North..., Lasiuk [/bib_ref]. Unlike virtually all other psychiatric conditions, which can be diagnosed solely on the basis of whether symptoms are present or absent, a diagnosis of PTSD requires exposure to a traumatic event or events. It often exists concurrently with other psychiatric disorders [bib_ref] Posttraumatic stress disorder in the National Comorbidity Survey, Kessler [/bib_ref] , making the relative contributions of each respective disorder to functional impairment potentially hard to discern. In medically ill populations, symptoms of PTSD are frequently expressed in nuanced and highly idiosyncratic ways and may not be captured through simple self-report questionnaires [bib_ref] Posttraumatic stress disorder following medical illness and treatment, Tedstone [/bib_ref] [bib_ref] Posttraumatic stress disorder in older adults: a conceputal review, Averill [/bib_ref] [bib_ref] Post-traumatic stress disorder in the community: an epidemiological study, Davidson [/bib_ref]. Additionally, self-report measures typically do not allow researchers to determine whether a constellation of symptoms reflect PTSD or a time-limited adjustment disorder [bib_ref] Posttraumatic stress disorder following medical illness and treatment, Tedstone [/bib_ref]. For these and other reasons, the accurate identification of PTSD or PTSS in time-limited research contexts is a significant challenge. Although many investigations of PTSD following critical illness have used methodological rigor, the existing body of work on the subject has a number of significant limitations, as is often the case with early explorations in most arenas. These limitations raise questions about the prevalence rates of PTSD and the magnitude of the problem that PTSD represents to ICU survivors.
## Limitations of existing studies
As previously described, the methodological limitations of the aforementioned studies are significant and may have contributed to overestimates of PTSD or PTSS prevalence. In particular, the practice of using screening tools for diagnostic purposes is problematic. Certainly, screening tools and questionnaires vary widely in quality and comprehensiveness, and some self-report questionnaires possess fairly robust psychometric properties [bib_ref] Systematic review of screening instruments for adults at risk of PTSD, Brewin [/bib_ref]. Nevertheless, such instruments are not typically intended to definitively identify the presence, absence, or severity of PTSD and tend to yield significantly higher false-positive rates than comprehensive diagnostic measures such as the SCID-PTSD and the Clinician-Administered PTSD Scale [bib_ref] The development of a Clinician-Administered PTSD Scale, Blake [/bib_ref] , although this is not always the case.
A study of burn survivors conducted by Tedstone and Tarrier [bib_ref] An investigation of the prevalence of psychological morbidity in burn injured patients, Tedstone [/bib_ref] may be instructive in this regard as it showed that whereas nearly 40% of their cohort were classified as 'PTSD cases' via the IES, only 2% were found to actually have PTSD when assessed with a comprehensive instrument, the Penn Inventory. Additionally, most screening tools have not been validated on patients with critical or life-threatening illness, thus responses to various questions may be confounded (for example, anticipating a 'foreshortened future' may be related to the experience of suffering from a particular medical condition and not a symptom of anxiety) [bib_ref] Posttraumatic stress disorder following medical illness and treatment, Tedstone [/bib_ref] [bib_ref] A systematic and conceptual review of posttraumatic stress in childhood cancer survivors..., Bruce [/bib_ref]. Additionally, few screening tools assess DSM-IV criteria A (exposure to a traumatic stressor) and F (the presence of clinically significant impairment), although the positive endorsement of both criteria must occur for PTSD to be diagnosed. The failure to assess criteria A and F is problematic, particularly because the symptoms of PTSD reported by individual ICU survivors (and attributed to an episode of critical illness by researchers) could potentially be the result of exposure to prior traumatic stressors.
Although some may argue that critical illness and associated factors such as prolonged hospitalization and mechanical ventilation are always traumatic stressors, this is not necessarily the case; the degree to which these events are experienced as traumatic may be mediated by age, severity of illness, abruptness of onset, religious faith, and individual interpretation [bib_ref] Psychopathology following trauma: the role of subjective experience, Creamer [/bib_ref]. Among individuals who neither experience an acute emotional response nor interpret a potential stressor as extremely disturbing and frightening, the likelihood of developing PTSD is very low [bib_ref] Psychopathology following trauma: the role of subjective experience, Creamer [/bib_ref] [bib_ref] Trauma and posttraumatic stress disorder in the community, Breslau [/bib_ref] [bib_ref] A dual representation theory of posttraumatic stress disorder, Brewin [/bib_ref] [bib_ref] Fear, helplessness, and horror in posttraumatic stress disorder: investigating DSM-IV criterion A2..., Brewin [/bib_ref] [bib_ref] A cognitive model of posttraumatic stress disorder, Ehlers [/bib_ref].
In addition to relying primarily on screening tools, a majority of investigations failed to assess for previous or intervening trauma, although such information is highly relevant in determining both the genesis of PTSD symptoms and the unique contributions of ICU treatment to the development of PTSD. Data suggest that a majority of community-dwelling individuals have been exposed to at least one traumatic event during their lifetime [bib_ref] Trauma exposure and post-traumatic stress disorder in the general population, Frans [/bib_ref] and that those individuals with chronic diseases such as HIV, diabetes, and musculoskeletal disorders (conditions common among ICU cohorts) have unusually high levels of trauma exposure [bib_ref] Posttraumatic stress disorder and physical illness: results from clinical and epidemiologic studies, Boscarino [/bib_ref] [bib_ref] Traumatic stress in HIV-infected women, Kimerling [/bib_ref] [bib_ref] Relationships among trauma exposure, chronic posttraumatic stress disorder symptoms, and self-reported health..., Kimerling [/bib_ref]. Whether the PTSD symptoms endorsed in the studies to date are primarily a function of ICUrelated events or instead are influenced by other traumatic exposures is a crucial question, but one that (in part due to the limitations of current research) cannot be answered.
Yet another limitation of research on PTSD and critical illness pertains to sampling issues. In studies of PTSD in more established populations (that is, combat survivors, victims of sexual assault, and patients with cancer), sample sizes are often quite large and patients are in many cases relatively homogenous. In contrast, the largest study of PTSD following critical illness contained fewer than 150 patients at follow-up, and the majority of investigations consisted of fewer than 50 patients at follow-up and included patients with substantial differences with regard to key characteristics, including the time to PTSD assessment. These issues, along with consistently and strikingly low follow-up rates, raise questions about the generalizability of study findings and the degree to which study participants are representative of typical critically ill populations. It may be, for example, that high-functioning ICU survivors without psychological sequela might conclude that the study participation is of little value to them and thus decline, or that subjects with PTSD might be particularly inclined to participate as a way of seeking help. Alternatively, it may be that some ICU survivors with PTSD may be less likely than their ICU counterparts to participate because the intense emotional distress they experience precludes them from doing so.
## Critical illness as a traumatic stressor
Although the experience of critical illness is undoubtedly stressful, aspects of this experience differ in nature from more traditionally defined and widely studied 'traumas' such as severe burns, automobile accidents, sexual assaults, and exposures to combat. For example, ICU patients are frequently unaware of the degree of life-threat their illness poses until after the illness is largely resolved. Additionally, the development of critical illness is frequently a continuation or acceleration of a longstanding disease process (for example, patients with chronic obstructive pulmonary disease have an exacerbation of symptoms, necessitating ICU care) as opposed to an abrupt occurrence. Despite these caveats, key factors associated with critical illness may be traumatogenic. These could potentially include the diagnosis of critical illness, the unique stresses often associated with ICU care such as intubation and weaning from mechanical ventilation, and the occurrence of nightmares and delusions. The cumulative effects of these factors could increase the likelihood of developing PTSD, particularly in patients with pre-existing vulnera-bilities such as a prior history of trauma exposure or a history of chronic medical illness [bib_ref] Lifetime traumas and mental health: the significance of cumulative adversity, Turner [/bib_ref] [bib_ref] The experience of chronic illness and post-traumatic stress disorder: the consequences of..., Alonzo [/bib_ref] [bib_ref] Cumulative adversity and posttraumatic stress disorder: evidence from a diverse community of..., Lloyd [/bib_ref] [bib_ref] The need for mental health services research focusing on poor young women, Miranda [/bib_ref].
As others have observed, altered mental status (in the forms of both delirium and coma) is common in the ICU, raising important questions about the role of memory (that is, the ability to remember traumatic events) in mediating the development of PTSD [bib_ref] Improving recognition of delirium in the elderly, Foreman [/bib_ref]. The importance of specific explicit memories (memories pertaining to facts and events, which are accessible to consciousness) [bib_ref] Human memory, Parkin [/bib_ref] in the generation and maintenance of PTSD is difficult to overestimate as they are the basis for nightmares, flashbacks, and intrusive thoughts and contribute to symptoms of avoidance and re-experiencing. Current evidence suggests that the absence of episodic memory for a traumatic event is protective against the development of PTSD; a majority of studies have shown that the risk of PTSD is markedly lower in individuals unable to recall a traumatic event than in those with explicit memory for such an event (or events) [bib_ref] Mild traumatic brain injury does not produce post traumatic stress disorder, Sbordone [/bib_ref] [bib_ref] The long term psychiatric consequences of accidental injury, Malt [/bib_ref] [bib_ref] Acute and chronic posttraumatic stress disorder in motor vehicle victims, Ursano [/bib_ref] [bib_ref] Do patients with mild brain injury have post-traumatic stress disorder too?, Bontke [/bib_ref]. However, some contemporary theories suggest that PTSDcan develop in patients with impaired consciousness for the following reasons: (a) patients can experience the traumatic event after they regain consciousness, (b) processing occurs at an implicit level during periods of impaired consciousness (that is, due to psychological distress encoded by amygdala activation, re-experiencing of symptoms can occur with any memory of the event), and (c) some people appear to reconstruct memories or experiences from photographs, reports that then 'become memories' that may provide the basis for the generation of PTSD symptoms even in the absence of conscious awareness [bib_ref] A dual representation theory of posttraumatic stress disorder, Brewin [/bib_ref] [bib_ref] Posttraumatic stress disorder and traumatic brain injury: can they co-exist?, Bryant [/bib_ref] [bib_ref] A cognitive neuroscience account of posttraumatic stress disorder and its treatment, Brewin [/bib_ref] [bib_ref] Implicit memory: a selective review, Schacter [/bib_ref].
# Conclusion
The relationship between critical illness and PTSD has been assessed in a limited number of studies over the last decade and a half. These studies have varied widely in their aims and methodological rigor but have raised awareness and generated valuable data and important insights. For example, we now recognize that sedation strategies can influence the development of PTSD symptoms. Additionally, more recent evidence suggests that individuals with predominantly factual, as opposed to delusional, recollections of the ICU may be at reduced risk for PTSD. Furthermore, it appears that the presence of premorbid mental health problems increases the likelihood of developing PTSD in survivors of the ICU.
Despite the growing recognition that PTSD may occur following an episode of critical illness, the extent to which it can reliably be considered a threat is unknown, due to the methodological limitations and conflicting results of the current studies. It is highly probable that investigations to date have tended to overestimate PTSD prevalence because of an over-reliance on screening tools (as opposed to diagnostic tools), questionable interpretations of available data, the lack of evaluation of non-ICU-related causes of PTSD, low followup rates, and other significant limitations. It is worth noting, in this regard, that the three studies reporting the highest rates of actual PTSD (>50%) had sample sizes of between 11 and 27 patients. Developing conclusions about prevalence on the basis of such limited investigations is both extremely imprudent and inconsistent with sound scientific practice. Nevertheless, PTSD clearly occurs and persists in a subset of ICU survivors.
Continued investigation of PTSD in critically ill populations is vitally important for determining the nature and scope of the problem and evaluating possible interventions. However, the relevance and value of a program of investigation will be limited unless it employs the same methodological rigor that characterizes the study of PTSD in other better-established populations such as combat veterans and cancer patients. To that end, specific guidelines should be adhered to and specific goals aggressively pursued. First, studies focused on PTSD as an outcome should use appropriate diagnostic tools and should focus not only on the identification of symptoms but also on the assessment of clinical significance. Researchers should attempt to use populations sufficiently large and representative so as to determine the approximate prevalence of PTSD in critically ill cohorts. In addition to evaluating prevalence rates, investigators should study rates of symptom remission. Second, the incidence of other potentially relevant historical or intervening traumatic stressors and trait variables (for example, neuroticism and anxiety) should be explored. Third, studies should more fully explore the specific etiologies of ICU-related PTSD, placing particular emphasis on the contributions of factual versus delusional memories to the development of PTSD. Fourth, studies should examine the effects of sedation strategies on the development of PTSD, focusing on the identification of strategies that may be protective against the development of PTSD. Finally, studies should assess specific risk factors for the development of PTSD in ICU survivors, focusing in particular on the identification of modifiable risk factors and potential interventions that might reduce the incidence of PTSD or PTSD symptoms. Understanding the nature of the relationship between critical illness and PTSD is a challenge that demands attention, particularly in an era when mental health professionals are beginning to recognize the significant and sometimes profound costs (interpersonal, vocational, medical, and financial) associated with this psychiatric syndrome.
## Key messages
- PTSD or PTSD symptoms are reported to occur in between 5% and 63% of ICU survivors, and key risk factors include duration of hospital and ICU stays, duration of ventilation, pre-existing psychiatric history, and the presence of delusional memories.
- Reported rates of PTSD prevalence following the ICU tend to be extremely high relative to other trauma populations, including medical and surgical patients, and are likely to be overestimates.
- Studies of PTSD following critical illness are characterized by significant methodological shortcomings, which raise key questions about the actual prevalence rates of PTSD and the generalizability of study findings.
- Future studies on PTSD should be more methodically rigorous and should use larger and more homogeneous samples while also employing comprehensive diagnostic, as opposed to screening, instruments.
[fig] a: Definition obtined from the Dignostic nd Sttisticl Mnul of Mentl Disorders, Fourth Edition (DSM-IV). [/fig]
|
Vasodilator agents improve hemodialysis vascular access patency
Vascular access (VA) failure is an important problem for patients undergoing hemodialysis, and maintaining VA patency is challenging. In this study, we used a nationwide database to investigate the effects of nitrate, as a vasodilator, on VA failure in hemodialysis patients.We investigated the Korean insurance claims data of hemodialysis patients who underwent angioplasty for VA failure between January 2012 and December 2017. The patients were divided into 2 groups: those not receiving vasodilator therapy (controls) and those receiving any vasodilator administration (vasodilator treatment, VDT). The primary endpoint was VA primary patency, defined as the time between arteriovenous dialysis access creation and the first percutaneous transluminal angioplasty (PTA).During the study period, a total of 6350 patients were recruited, 409 (6.4%) patients assigned to the VDT group and 5941 (93.6%) controls. PTA was performed in 998 patients (15.7%), including 8 in the VDT group and 990 controls. The VA site PTA rate was significantly lower in the VDT group (2.0%) than in the control group (16.7%, P < .001). In the subgroup analysis, the patency rates associated with the different vasodilators were similar (P = .736). All vasodilators, except molsidomine, improved the patency rate by approximately 20%.In this large national database study, vasodilator administration was associated with higher VA primary patency, compared with controls, in hemodialysis patients. VDT may have a beneficial effect on maintaining VA patency in patients undergoing hemodialysis.Abbreviations: AVF = arteriovenous fistula, AVG = arteriovenous graft, ESRD = end-stage renal disease, HIRA = the Health Insurance Review and Assessment Service, PTA = percutaneous transluminal angioplasties, VA = vascular access, VDT = vasodilator treatment.
# Introduction
Vascular access (VA) is essential for patients with end-stage renal disease (ESRD) undergoing hemodialysis. The survival and quality of life of these patients depend on the adequacy of dialysis via a well-functioning VA.However, creating and maintaining a well-functioning VA is a critical challenge. A VA can be obtained using a native (autologous) arteriovenous fistula (AVF), arteriovenous graft (AVG) connecting artery and vein with a prosthetic graft, or a central venous catheter,with most patients undergoing hemodialysis receiving an AVF or AVG. After VA creation, the shear stress on venous endothelial cells, due to increasing flow, induces venous vascular remodeling and vascular dilatation, leading to VA maturation.However, if only vascular remodeling occurs, without dilatation, myofibroblasts are activated and infiltrate into the intima and differentiate into smooth muscle cells.After the proliferation and migration of smooth muscle cells, the AVF endothelial intima may develop neointimal hyperplasia, leading to AVF stenosis and dysfunction.Thus, venous stenosis at the vein-artery anastomosis, in AVF, or at the vein-graft anastomosis, in AVG, is the main cause of VA failure.Nitrates and molsidomine, nitric oxide group-containing vasodilators, affect dilatation of systemic and coronary vascular beds; nicorandil, another vasodilator, causes arterial, and venous dilatation.Although these vasodilators have demonstrated systemic or coronary artery vasodilation effects, there is limited evidence about their effects on VAs. Thus, we investigated a large, nationwide, population-based database to determine the effects of vasodilators on VA patency.
# Methods
## Study design and participants
This retrospective cohort study used information from the National Health Insurance Service-National Health Screening Cohort and the Health Insurance Review and Assessment Service (HIRA). The National Health Insurance Program covers nearly 100% of the Korean population, and the NHIS-HEALS is a representative sample cohort that was randomly selected from the total eligible Korean population, for 1 year; we merged the oneyear NHIS-HEALS data sets from 2012 to 2017. Further, the HIRA data, such as age group, sex, comorbidity, prescription information, surgical history, and treatment procedure(s), were included. The cohort's representation of the general population has been previously described.
## Data collection
The International Classification of Diseases, 10th revision, codes were used to identify patients >19 years old, with newly diagnosed ESRD (N185 or N189) and undergoing hemodialysis (N189 and V001). Thereafter, we selected patients who underwent VA operations, using procedure codes such as AVF or AVG, from 2012 to 2017; the specific procedure codes examined were: O2011 (external arteriovenous shunt for hemodialysis), O2012 (internal arteriovenous shunt for hemodialysis), O2081 (fistula formation-autologous vein (hemodialysis)), and O2082 (fistula formation-artificial vein (hemodialysis)).
In addition, we extracted the data for patients who underwent percutaneous transluminal angioplasties (PTAs), including the diagnosis codes T823 (mechanical complication of other vascular grafts), and M6597 (percutaneous transluminal angioplasty (others)), with the claim dates defined as the procedure dates.
Vasodilator prescription data were extracted from the HIRA. The vasodilator agents included isosorbide mononitrate, isosorbide dinitrate, molsidomine, and nicorandil (detailed medication codes are included in, Supplemental Digital Content, http://links.lww.com/MD2/A537). According to the data, the patients were assigned to either the vasodilator treatment (VDT) or control group.
## Study outcomes
The primary endpoint was the VA primary patency rate after the AV shunt operation. The period from the AV operation day to the first PTA event was defined as the primary patency duration of the AV shunt. Primary endpoint was compared according to VDT.
# Statistical analysis
Continuous variables are presented as means with standard deviations; categorical variables are presented as numbers with percentages. Categorical variables were analyzed using the x 2 test or Fisher exact test. Differences between the VDT and control groups were analyzed using Student t-test. The observed patency rate was estimated using the Kaplan-Meier method. A Cox proportional hazards model, which included all of the baseline variables and vasodilators for the between-group comparison, was used to determine the relationship between the clinical variables and primary patency. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated according to sex (male), older age (>65 years old), past medical history (hypertension, diabetes, myocardial infarction, dyslipidemia, stroke, and peripheral artery disease), and medication history (particularly aspirin, clopidogrel, and statin use). Furthermore, we analyzed and compared VA patency rate among each vasodilator. All statistical analyses were performed using R, version 4.0.2 (R Foundation for Statistical Computing, Vienna, Austria; http:// www.R-project.org) and SAS (SAS Enterprise Guide 7.1; SAS Institute, Cary, NC, USA). P value <.05 was considered statistically significant.
# Data sharing statement
HIRA data are third-party data not owned by the authors. HIRA data are available upon visit or by mail upon direct, email or fax submission of the data set request form and declaration of data use that is downloadable from the "HIRA" website (http://www. hira.or.kr/dummy.do?pgmid=HIRAA070001000450) and upon payment of the transfer of data request fee (300 000 KRW per data set).
## Ethics approval
The study protocol was approved by the Institutional Review Board at Hallym University Kangnam Sacred Heart Hospital (approval number: 2019-12-015-001), and all procedures were performed in accordance with the tenets of the Declaration of Helsinki.
# Results
## Baseline characteristics
A total of 6350 patients (mean age, 69.9 years; 3819 (60.1%) males) were included in this study. As shown in, 409 (6.4%) patients were assigned to the VDT group and 5941 (93.6%) comprised the control group. The patients in the VDT group were older than those in the control group. In the VDT group, the proportions of patients with histories of diabetes mellitus, myocardial infarction, dyslipidemia, and stroke were higher than in the control group; however, hypertension was more prevalent in the control group. In addition, aspirin, clopidogrel, and statins were prescribed significantly more frequently in the VDT group.
## Vascular access patency rate
During the study period, 998 patients (15.72%) underwent PTAs. The incidence of PTAs involving the VA site was significantly lower in the VDT group (1.96%) than in the control group (16.66%, P < .001;. The VA one-month patency rates were similar between the 2 groups; however, in the Kaplan-Meier analysis, the 1-year patency of the hemodialysis VA site was significantly higher in the VDT group.
## Related factors and subgroup analysis
To compare patency rates in patients prescribed different vasodilators, a subgroup analysis was performed. When baseline variables and the vasodilators were included in the univariate Cox regression analysis examining potential AV stenosis risk factors, the patency rates were similar for the different vasodilators. The patency rate for males was about 20% greater than that for females (HR, 0.80; 95% CI, 0.71-0.91; P = .001;.; all the vasodilators, except molsidomine, were associated with patency rate improvements of approximately 20%.
In the multivariate Cox regression analysis, a history of vasodilator administration was associated with a lower risk of PTA, compared with patients not receiving a vasodilator (HR, 0.15; 95% CI, 0.06-0.40; P < .001). Further, the PTA risk was significantly lower among men than among women. The use of aspirin and statins was not related to PTA risk; however, clopidogrel use was associated with a significantly increased PTA risk.
# Discussion
In this study, vasodilator use was associated with improved VA performance in patients undergoing hemodialysis. Although there were no differences in the VA patency rates among the examined vasodilator drugs, patients receiving vasodilators had fewer PTAs and higher VA patency rates than those not receiving these drugs. In addition, male patients and patients with histories of myocardial infarctions and strokes tended to have reduced PTA risks. However, clopidogrel use was associated with a higher PTA risk than was observed for other patients undergoing hemodialysis. Venous stenosis at the vein-artery anastomoses in AVFs or the vein-graft anastomoses in AVGs is the most common cause of VA failure. The venous stenosis is caused by vascular intimal hyperplasia and thrombosis, which is triggered by platelet activation, endothelial cell injury, and vascular smooth muscle cell proliferation.Every year, AV stenosis leads to AV dysfunction and causes 20% to 30% of hemodialysis patients to require inpatient treatment; it also results in higher patient mortality rates.Therefore, improving the prognoses of hemodialysis patients by preventing narrowing or blockage of the blood vessels that can lead to AV fistula dysfunction is important.To solve this problem, various interventional procedures, such as using drug-eluting balloons or stents, for reducing AV fistula stenosis are being performed; medical treatment for preventing stenosis and obstruction is also being considered.Systemic medical therapies involving the use of antiplatelet agents (aspirin and clopidogrel), omega-3 polyunsaturated fatty acids (fish oils), and statins reduce VA failure by promoting VA maturation and reducing stenosis and thrombosis through antiproliferative, antiaggregatory, anti-inflammatory, and vasodilatory effects.In our report, aspirin use was not associated with VA patency, similar to the results from other studies. For example, Kim et al reported a large-scale prospective study in which aspirin did not show a protective effect on vascular patency.In addition, the effect of clopidogrel for VA patency did not affect functional AV fistula significantly. Otherwise, the patients prescribed clopidogrel was associated with higher PTA risk in this study.A research documented that clopidogrel was more prescribed to the patients who have cardiovascular risk factors as postcoronary intervention therapy in the clinical field. We considered that clopidogrel might be more prescribed for patients with severe comorbidity, and that could affect the result.However, a systematic review reported that antiplatelet agents, including aspirin, prevented VA blockage.Thus, despite some studies showing the expected effects, these results are not universal. As a result, the effects of these drugs remain unclear and subject to debate.In this study, we documented that older patients, those with diabetes, and females had higher PTA risk than patients without these traits. Similar results have been shown in other studies; for example, Wen et al reported that female sex and advanced age were independent risk factors for VA dysfunction.Additionally, Smith et al reported that advanced age and the presence of diabetes are risk factors for VA dysfunction.Age might be associated with an increased PTA risk because of the higher incidence of comorbidities, such as peripheral vascular disease and diabetes, in patients of advanced age. The sex-associated difference in risk may be related to the diameter of women's arteries being smaller than those of men; therefore, lower AVF patency rates in women, compared with men, may be expected.However, a single-center, retrospective study revealed no differences in the construction of AVFs for patients in different age groups.In the case of sex, a meta-analysis study documented similar one-year patency rates, maturation, and vessel diameters between males and females.Therefore, age and sex, as independent factors of VA patency, continue to be debated.Relative to the impact of diabetes, a systematic review documented that primary patency rates appear to be lower in patients with diabetes than in those without the disease, similar to the results of our study.Although several studies regarding factors affecting VA patency have been reported, there are few studies describing the effects of vasodilators on VA. When an AVF is created, blood flow causes shear stress on the vascular wall and also triggers endothelial cells to produce nitric oxide, which is necessary for vasodilation.Endothelial nitric oxide synthase seems to be essential for vascular maturation and function, but its effect on AVF maturation has not been clearly established. Daniel et al reported that overexpression of the nitric oxide synthase system results in distinct hemodynamic and wall mechanical profiles associated with favorable AVF remodeling, in a mouse model. The authors documented that AVFs in mice overexpressing nitric oxide showed larger lumen areas, inducing smoother blood flow, lower wall shear stress, blood vorticity, inner wall circumferential stretch, and radial wall thinning at the anastomoses.We focused on lumen dilatation as a cornerstone to prevent VA stenosis, and analyzed the effect of vasodilators on VA patency in a large-scale study of patients undergoing hemodialysis. As suggested by the mouse model study, we assumed that vasodilator administration triggered dilation of the lumen diameter of the VA, leading to lower shear stress at the vascular wall. Thus, this phenomenon might be associated with the observed VA patency improvements in patients undergoing hemodialysis and being treated with vasodilators.
There are some limitations to our study. The major limitations of this study originate from the inherent features of the the National Health Insurance Service-National Health Screening Cohort cohort.The possibility of misclassification bias exists for the diagnosis of ESRD involving hemodialysis, which was defined using healthcare usage records. In addition, some data may be underestimated due to the following reasons. Second, most patients might have received vasodilators due to cardiovascular disease. If patients undergoing hemodialysis have underlying cardiovascular disease, the mortality rate is increasedand some patients might die before the PTA procedure. Since the outcome of this study focused on VA patency, using the PTA code, those who died before being diagnosed with VA stenosis could not be included. Third, we did not include surgical thrombectomy as a treatment for VA stenosis. Fourth, this study is a retrospective study based on NHIS codes. Therefore, we could not confirm the mechanism of vasodilation for VA. Finally, the number of patients was small and mortality data were not available for analysis, even though we enrolled all patients with ESRD and requiring hemodialysis who underwent VA creation during the study period. In addition, we missed the propensity scoring matching analysis due to technical and time limitations. However, the statistically significant improvement in VA patency in the VDT group might be considered to be a meaningful indication of the vasodilator effect.
In conclusion, vasodilator administration might help improve VA patency in patients undergoing hemodialysis. Thus, these results support a potential therapeutic approach involving the use of vasodilators. Moreover, the study provides evidence to support future research into the mechanism and role of these agents in the preservation of hemodialysis VAs. |
Fecal microbiota transplantation for ulcerative colitis: a prospective clinical study
Background: Fecal microbiota transplantation may contribute to disease remission in ulcerative colitis; however, the factors that determine the effects of treatment remain unknown. The aim of the present study was to prospectively investigate the clinical efficacy of fecal microbiota transplantation in patients with ulcerative colitis and identify the bacterial signatures associated with clinical remission.Methods: A total of 20 patients with ulcerative colitis were included in this prospective and uncontrolled study. All patients underwent gastroscopy five times, once every 3 weeks. Clinical indices were used to assess the efficacy of fecal microbiota transplantation, as well as the Mayo score, a score used to evaluate the extent of intestinal mucosal lesions in patients with ulcerative colitis. The changes in intestinal flora were detected by 16S ribosomal RNA-sequencing, and the relationship between ulcerative colitis and intestinal flora was analyzed.Results: After treatment, clinical index scores for diarrhea, abdominal pain, and blood stool decreased significantly (p < 0.05). Erythrocyte sedimentation rate and C-reactive protein levels had not changed significantly; however, the clinical index score for intestinal mucosal lesions and the Mayo score decreased significantly. In addition, 16S ribosomal RNA-sequencing revealed that the intestinal flora in patients diagnosed with ulcerative colitis was different from that of donors. Conclusion: Fecal microbiota transplantation has a potential therapeutic value for the treatment of ulcerative colitis as it changes the abundance of bacterial flora and improves the scores for diarrhea, abdominal pain, and mucous membrane lesions in patients with this disease. Trial registration: The clinical trial was retrospectively registered with ClinicalTrials.gov (NCT03016780) on January 11th, 2017.
# Background
Ulcerative colitis (UC) is a chronic and progressive intestinal inflammatory disease that can seriously affect patient quality of life. The main pathogenic mechanism of UC is thought to be aberrant activation of the immune system in response to a change in the gut environment. However, the cause of this pathological immune system activation is not fully understood. In recent years, a growing body of evidence suggests that intestinal microorganisms may play an important role in UC pathogenesis. The species diversity of intestinal flora in patients with UC differs from that of healthy subjects. For example, patients with UC exhibit decreased intestinal populations of members of the phyla Firmicutes and Bacteroidetes, and increased populations of Lactobacillus. In particular, the Desulfovibrio and Clostridium genera have been closely linked to UC. Thus, the development of UC is closely related to changes in the intestinal flora.
Healthy intestinal flora colonizes the intestinal mucosal epithelial cells and enhances the intestinal bio-barrier function. The flora adheres to the surface of the intestinal mucosa to form a chemical barrier against external stimuli, which regulates intestinal immunity. However, once the homeostasis of the intestinal microenvironment is disturbed, patients are susceptible to intestinal diseases.
A range of therapies are used in the treatment of UC, including 5-aminosalicylic acid, hormone therapy, immunosuppressants, biological agents, and surgery. However, these treatments have poor efficacy. Therefore, there is a requirement for new therapeutic strategies for the treatment of UC. Fecal microbiota transplantation (FMT) may be of therapeutic value in patients with UC by contributing to the repopulation of healthy intestinal flora. However, conflicting results have been reported regarding the efficacy of this treatment. This study evaluated clinical efficacy and safety of FMT and analyzed the relationship between UC and intestinal flora. Additionally, the effect of the intestinal flora on the intestinal mucosa was examined.
# Methods
## Study design
This prospective uncontrolled study was carried out at the Gastroenterology Department of The First Affiliated Hospital of Chengdu Medical College (Chengdu, People's Republic of China). Patients with UC were enrolled between July 2016 and October 2017.
## Study population
In this study, 20 patients who met the UC diagnostic criteria were recruited, according to typical clinical, endoscopic, and histopathological findings. A detailed history was taken from each participant, including smoking status, disease duration, medication, and history of previous intestinal surgery.
## Inclusion and exclusion criteria
Inclusion criteria were as follows: (1) Subjects voluntarily participated in the trial and signed an informed consent form; (2) Subjects were aged 18 to 75 years, both sexes were included; (3) Subjects met the diagnostic criteria for UC; and (4) Subjects were able to communicate well with the researcher and comply with the test requirements.
Exclusion criteria were as follows: (1) Subjects were pregnant or unable to give informed consent; (2) Subjects had used immunosuppressive agents in the past 6 months; (3) Subjects had suffered of severe immunodeficiency in the previous 6 months; (4) Subjects had taken antibiotics or probiotics within the previous 6 weeks;Subjects had serious complications, such as local stenosis, intestinal obstruction, intestinal perforation, toxic colon expansion, colon cancer, or rectal cancer; (6) UC was accompanied by a primary disease, such as a cardiovascular, cerebrovascular, hepatic, renal, or hematologic disease, or by a mental illness; and (7) The subject's condition was aggravated by cessation of their normal treatment. In these cases, emergency measures were taken, the efficacy could not be judged, or the data were incomplete.
## Donor selection
Donated stool for FMT was obtained from four donors, aged between 23 and 27 years. Donors had no diagnosed medical conditions that could be potentially associated with changes in gut microbiota. Donors who had taken antibiotics or probiotics within the previous month were not included for screening. All donors were required to complete a Donor Questionnaire form. In order to prevent transmission of infectious diseases from donor to recipient, all donors underwent stool test screening [(bacterial culture and identification; fecal flora ratio examination; human rotavirus antigen determination; parasite egg detection; microscopy; serologic tests (hepatitis A virus, hepatitis B virus, hepatitis C virus, HIV antibody, syphilis, herpes simplex virus, EB virus); and Cryptosporidium, Cyclospora, and Giardia antigen detection)].
Every donor provided stool samples for five patients. Patients were divided into four groups, and patients from every group received stool samples from the same donor. Clinical efficacy was compared among the four groups. Clinical remission and response rates were calculated for all subjects after treatment.
## Fmt procedure
Patients received the bowel lavage (polyethylene glycol 4000) for colonoscopy preparation the day before FMT. The median amount of donor feces was calculated (50 g) and used for FMT preparation. Donors were instructed to collect feces in a small container and to bring it to the hospital on the day of the scheduled transplant. A total of 250 mL extracted fecal suspension was prepared with 250 mL 0.9% NaCl using a conventional blender and was divided into 50 mL syringes. Filtered fecal microbiota suspension was administered into a catheter inserted into the duodenum by gastroscopy. After the procedure, the patient was returned to the ward, ensuring that their head stayed in a low position for 60 min. Following FMT, bowel movements were avoided for 0.5 h, food intake was prohibited for 1 h, and physical activity was prohibited for 2 h.
## Baseline patients' and healthy donors' characteristics
Patients' and healthy donors' baseline characteristics, including sex, age, height, weight, history of taking medicine before FMT for inflammatory bowel diseases, course of the disease, levels of inflammatory markers [erythrocyte sedimentation rate (ESR) and Creactive protein (CRP)], colonic mucosal score, and Mayo score before FMT, are listed in.
## Clinical index scores
The clinical symptom scores used to measure the efficacy of FMT were as follows: the diarrhea score, the abdominal pain score, the pus and blood stool score, the Mayo score, the bloody stool score, mucosal manifestation scoring, and colonic mucosal scoring. Diarrhea was evaluated as follows: 0 points, no diarrhea; mild diarrhea (< 4 times/day), 3 points; moderate diarrhea (4-6 times/day), 6 points; severe diarrhea (> 6 times/day), 9 points. Abdominal pain was evaluated as follows: no abdominal pain, 0 points; mild abdominal pain, 3 points; moderate abdominal pain (4-6 times/ day), 6 points; severe abdominal pain, 9 points. Pus and blood in stool were evaluated as follows: no pus and blood in stool, 0 points; mild pus and blood, 3 points; moderate pus and blood, 6 points; severe pus and blood, 9 points.
The Mayo score was used to evaluate the extent of intestinal mucosal lesions in UC. The scoring was as follows: normal stool frequency, 0 points; more than normal stool frequency (1-2 times/day), 1 point; more than normal stool frequency (3-4 times/day), 2 points; very high (more than 5 times/day), 3 points.
Bloody stool was evaluated as follows: no blood in the stool, 0 points; a little blood in the stool, 1 point; obvious bloody stool, 2 points; mostly bloody stool, 3 points. The mucosal manifestation scoring was categorized as follows: normal mucosa, 0 points; mild fragility, 1 point; moderate fragility, 2 points; moderate fragility with exudation, 3 points. The colonic mucosal scoring criteria were as follows: normal intestinal mucosa, 0 points; mucosal congestion and blood vessel blushing, 1 point; mucosal contact bleeding, 2 points; mucosal spontaneous bleeding, 3 points; and mucosal ulcers, 4 points.
Finally, the ESR and CRP levels were used as indicators of inflammatory reactivity. An automatic ESR analyzer was used to detect ESR, and transmitted immunoturbidimetric method was used to determine CRP at the laboratory of the First Affiliated Hospital of Chengdu Medical College.
# Intestinal flora analysis
16S ribosomal RNA sequencing (16S rRNA-seq) analysis was performed on the bacterial rRNA from stool of healthy donors and patients with UC before treatment and after the first and second treatment (groups d0, d1, and d2). DNA was first extracted from the stools of healthy donors and patients. DNA pre-amplification and sequencing was carried out by Tianjin Novo Zhiyuan.
First, PCR pre-amplification was conducted to determine whether the samples met the quality control criteria. The DNA primer sequences were 341-F: (5′-CCTACACGACGCTCTTCCGATCTN-3′) and 805-R: (5′-GACTGGAGTTCCTTGGCACCCGAGAATTCCA-3′). Barcode-tagged primers were used to perform PCR amplification of the 16S V3-V4 region. The products were subjected to quality inspection, purification, and library construction. After the alignment, 16S rRNAsequencing was performed on the Illumina Hiseq-PE250 technology sequencing platform, using the double-end sequencing method, resulting in 3,043,659 highquality sequences. Next, the relevant statistical analyses were performed.
# 16s rrna-seq analysis
After removal of the chimeras, the filtered highquality sequences were grouped into 8650 operational taxonomic units (OTUs). In all samples, 99.962 and 67.5% of the total sequence were assigned to 14 and 142 genera, respectively. Unclassified bacteria accounted for approximately 0.038% of the total sequence.
# Statistical analysis
Measurement indicators are represented as mean and standard deviation, and counting indicators are presented as number and percentage of each category. The intra-group comparison of measurement indicators was performed using paired t-test or a Wilcoxon's signedrank test. The count index was tested using a paired chisquared (χ 2 ) test, and the grade index was tested using a Wilcoxon's signed-rank test. P-values < 0.05 were considered statistically significant.
# Results
## Patient characteristics
This study included 11 men and 9 women aged from 18 to 73 years. All patients completed five rounds of FMT treatment, once every 3 weeks.
## Clinical outcomes
The diarrhea score showed a downward trend after treatment, as shown in(3.75 ± 3.49 before treatment; 0.79 ± 1.69 after the fifth treatment). The mean diarrhea score was significantly decreased following five rounds of FMT, as compared to the diarrhea score before treatment.
The abdominal pain score also showed a significant downward trend after treatment, as shown in(2.55 ± 2.63 before treatment; 1.42 ± 1.54 after the fourth treatment). The bloody stool score also showed a downward trend after treatment, as shown in(3.30 ± 2.36 before treatment; 0.79 ± 1.36 after the fifth treatment). The endoscopic intestinal mucosal score and the Mayo score were used to evaluate the intestinal mucosa and disease activity, respectively, in patients with UC before and after FMT. Post-treatment, the intestinal mucosal score (1.37 ± 0.60) was significantly lower than that at the pretreatment time point (1.80 ± 0.70, p < 0.05). Likewise, the post-treatment Mayo score (3.00 ± 2.00) was significantly lower than the pre-treatment Mayo score (5.00 ± 2.75)and. Compared with pre-treatment, The aspect of the intestinal mucosal under enteroscopy improved and the infiltration of inflammatory cells in the intestinal mucosa decreased after FMT.
The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were used as indicators of inflammatory reactivity. These measures did not change significantly following FMT treatment.
## Safety evaluation
No patient had serious adverse reactions during the study period and follow-up period. One patient developed skin erythema on the first day after treatment. We consider that it is related to allergic reaction, the erythema disappeared quickly after anti-allergy treatment. Other three patients had mild abdominal distension that resolved without clinical treatment within 24 h.
## Analysis results of intestinal flora
Relative abundance of intestinal flora in patients with UC and healthy donors ANOVA and LSD statistical analyses revealed no significant difference in intestinal microbial composition in patients with UC versus healthy donors at all intestinal. Although no significant differences in gut microbiota composition were found at the phylum level, a significant change in the composition of the flora was detected in patients with UC compared to healthy donors. We analyzed the composition of intestinal flora before treatment (d0) and after the first (d1) and second (d2) treatments, and found that at the level of the phylum, the proportion of Firmicutes in the d0 stage was higher, accounting for 54.0% of the total sequence reads, which was similar to the proportion of the donor group. The proportion of Firmicutes showed a downward trend after treatment in the group of patients with UC. Bacteroidetes in the donor group accounted for 41.8% in the healthy donor group, and accounted for 30.5, 33.4, and 37.8% in the d0, d1, and d2 groups, respectively. This gradual upward trend indicates that the relative abundance of Bacteroidetes gradually approached that of the donor group after treatment. Furthermore, the relative abundance of Proteus in the d0, d1, and d2 groups was significantly higher than that of the healthy donor group, accounting for 13.5, 15.3, and 16.0%, respectively (as shown in. At the genus level, the proportion of Bacteroides in the d0, d1, and d2 groups was 18.9, 15.6, and 20.2%, respectively; being lower than in the healthy donor group. Prevotella proportions in d0, d1, and d2 groups was lower than that of donor group and accounted for 8.6, 14.2, and 14.5% of sequence reads, respectively. Klebsiella abundance in the d0, d1, and d2 groups was 5.6, 5.9, and 1.2%, respectively, while that of the donor group was only 0.2%. In the d0, d1, and d2 groups, Streptococcus accounted for 1.8, 2.4, and 0.4%, respectively, compared to that of the donor group, which was 0.3%. The relative abundance of Streptococcus before treatment was significantly higher than that of the healthy donor group and decreased significantly in the d2 group (as shown in.
## Venn diagram analysis results
The Venn diagram reflects the overlap between OTU at different treatment stages, with a large number of overlapping OTUs in each treatment stage. A total of 1342 OTUs were found. The most highly contrasting OTU was that between the patient group d0 and the healthy donor group. However, following treatment, there was an increase in the number of overlapping strains and a decrease in the number of unique OTUs between the healthy donor group and d1 and d2 groups.
## Diversity analysis results of intestinal bacterial populations
Alpha diversity is used to measure the species diversity of a single ecological sample of a community, and is a comprehensive indicator reflecting the richness and uniformity of a population. Alpha diversity accounts for the following factors: observed species index, the Chao I index, and the Shannon index. A dilution curve of goods-coverage reflects whether the sequencing results are an accurate representation of the sample.shows the alpha diversity dilution curve of each sample. The dilution curve of goods-coverage is close to the plateau stage, indicating that the sequencing amount of this test is close to saturation. This indicates that the number of sequencing reads closely reflects the diversity After the fifth treatment 1.37 ± 0.60* 3.00 ± 2.00* *P < 0.05, the difference was statistically significant composition of the fecal flora of each sample in this experiment. The sequencing depth is sufficient. Judging by the OTU number, the Shannon index and the Chao I index, the bacterial diversity in the stool samples of patients with UC decreased with the number of treatments. However, this decrease was not statistically significant. The change in the bacterial composition of the stool samples was related to altered bacterial colonization of the intestine. Beta diversity reflects the similarity of microbial communities. The bacterial community clustering remained unchanged after FMT in patients with UC.
## Lda effect size analysis results
LDA Effect Size (LEFSe) is used to analyze species that have significant differences in abundance between groups (biomarkers). As shown in , a total of 58 significantly different genera were identified between the donor healthy group and the group of patients with UC by LEFSe. Among them, the intestinal bacterial populations of the healthy donor group were dominated by Lachnospiraceae, Ruminococcus, Parabacteroides, Sutterella, and Akkermansia, and .
# Discussion
This study compared changes in abdominal pain scores, diarrhea scores, bloody stool scores, endoscopic intestinal mucosal score, and Mayo scores before and after FMT treatment in patients with UC and evaluated the clinical efficacy of FMT in the treatment of UC. The results showed that the patients' abdominal pain score, diarrhea score, bloody stool score, intestinal mucosal lesion, and Mayo score significantly decreased after treatment. This is consistent with previous research results. However, CRP level and ESR did not significantly change following FMT treatment. This suggests that FMT can improve the clinical symptoms and mucosal lesions, reduce disease activity, and slightly reduce the disease inflammatory response in patients with UC.
The earliest application of FMT for the treatment of UC was attempted in 1988, and showed that patients' UC symptoms significantly improved. Brandt et al.followed up 6 patients with UC after FMT treatment and also found that their symptoms were alleviated. Kump et al.found that FMT improves the clinical symptoms of patients with UC by regulating intestinal flora. Our results also showed that FMT improves abdominal and bowel discomfort symptoms in patients with UC.
The intestinal microenvironment plays an important role in maintaining the intestinal mucosal immunity and regulating the intestinal function. Here, we found that in both patients with UC and healthy donors, intestinal flora was mainly composed of Bacteroidetes, Firmicutes, and Proteobacteria. However, the ratio of Bacteroides to Proteobacteria was significantly different between patients with UC and healthy donors, which is consistent with the findings of previous reports.
At the genus level, the relative abundance of Prevotella before treatment was lower than that of the donor group. The relative abundance of Klebsiella and Streptococcus was higher than that of the donor group. After treatment, the relative abundance of these three genera gradually became similar to that of the health donors. Therefore, the decrease of Prevotella, and the increase of Klebsiella and Streptococcus proportions may be important factors leading to the onset of UC.
Indeed, LEfSe analysis indicated that there was a difference in the intestinal flora between donor and patients. Samples from the d0 group of patients with UC were dominated by Klebsiella, Megamonas, Erysipelotrichaceae, Epulopiscium, and Dorea. These genera may be related to the pathogenesis of UC, and this information may therefore be of clinical value for improving the diagnosis of UC.
OTU-based Venn diagram analysis and found that the number of overlapping OTUs between patients with UC in different treatment stages and donor groups increased gradually, indicating a gradual return of the patient intestinal flora to the healthy state. Furthermore, the number of non-overlapping OTUs gradually decreased. This indicates that FMT can, to a certain extent, correct UC-associated dysbiosis. Due to the clinical efficacy of FMT treatment, it can be speculated that these dominant bacteria may improve the symptoms of patients with UC, but this hypothesis requires further verification.
Our findings showed that FMT is a safe and effective treatment for UC. After transplantation, the symptoms of diarrhea, abdominal pain, and bloody stools improved. Intestinal mucosal lesions improved, and the Mayo score decreased. Furthermore, we have shown that FMT can regulate intestinal flora. Therefore, FMT can be used as a novel therapy for the treatment of UC. However, if used widely in a clinical setting, the FMT procedure must be standardized (e.g., donor selection, stool preparation, delivery route, and dosing). Therefore, there is a requirement for further evidence from long-term and randomized controlled clinical studies examining donor and recipient microbiota composition. Our study has some limitations. First, the number of patients that we have recruit is not too much;Second, the study performed in a single institution,this may limite the Diversity of study. So, other more perfect study is necessary in the futuer.
# Conclusion
Fecal microbiota transplantation improves symptoms in patients with UC through changing the abundance of bacterial flora. This study provides a valuable treatment modality for UC.
# Funding
The present study was supported by the Science and Technology Project of The Health Planning Committee of Sichuan (grant no. 17ZD012), the Foundation of the Medical Association of Sichuan Province (grant no. S16019), and the Foundation of the First Affiliated Hospital of Chengdu Medical College (grant no. ZYFY2018XH01, grant no. ZYFY2017XH02, grant no. ZYFY2017XH03, grant no. ZYFY2017ZD02).
## Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
## Ethics approval and consent to participate
The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Chengdu Medical College (2017009) and the trial was registered with America Clinical Trials Registry (Clinical Trials: NCT03016780). All patients and donors gave written informed consent after counseling about the study and its potential risks and benefits.
## Consent for publication
Not applicable. |
Thermotherapy for knee osteoarthritis
Background: Osteoarthritis of the knee is one of the leading causes of pain and disability among adults. Thermotherapy has been widely used to treat knee osteoarthritis. But its efficiency has not been scientifically and methodically evaluated. The aim of this study is to assess the benefits of thermotherapy for people with osteoarthritis of the knee, in terms of pain, stiffness, and physical dysfunction.Methods: Eight databases will be searched from their inception to September 2020. They are as follows: PubMed, Embase, Cochrane Library, ClinicalTrials.gov, China Knowledge Resource Integrated Database (CNKI), Weipu Database for Chinese Technical Periodicals (VIP), Chinese Biomedical Literature Database (CBM), and Wanfang Database. Two researchers will independently select studies, collect data, and assess the methodology quality by the Cochrane risk of bias tool.Results: The systematic review will provide high-quality evidence to assess the benefits and harms of thermotherapy for people with osteoarthritis of the knee, in terms of pain, stiffness, and dysfunction of knee joint, and quality of life, as well as adverse events.Conclusion: The systematic review will provide evidence to assess the effectiveness and safety of thermotherapy for knee osteoarthritis patients.INPLASY registration number: INPLASY202140038.Abbreviations: KOA = knee osteoarthritis, RCTs = randomized controlled trials.
# Introduction
Knee osteoarthritis (KOA) is one of the leading causes of pain and disability among adults.Its prevalence has steeply increased during the last 6 decades; approximately 10% of the world's population aged 60 or older have symptomatic osteoarthritis of the knee.Globally, hip and knee osteoarthritis rank among the top 20 contributors to disability of all health conditions.Typical clinical features include pain, swelling, stiffness, and limited range of motion of the knee joint, often accompanied by a substantial decrease in quality of life. There is no cure for OA at present, and so objectives of management of symptoms of OA of the knee are to lessen pain and stiffness, maintain or improve mobility, and minimize disability.Treatment options include pharmacologic intervention, exercise therapy, surgery, and hot and/or cold therapy. Different physiotherapy treatments have been shown to help improve clinical symptoms and function of knee OA with fewer adverse effects than medical treatment. Thermotherapy is 1 such noninvasive therapy.
Thermotherapy is the application of heat to the body resulting in increased tissue temperature.Techniques for thermotherapy include the application of moxibustion, hot packs, superficial heat, and via diathermy (application of electromagnetic energy). Thermotherapy is used in rehabilitation to reduce pain and stiffness, and to increase mobility.Thermotherapy helps to relax muscles and increase circulation to the affected area, thus reducing pain and stiffness, although there is some concern that this may, in turn, worsen inflammation and edema. Thermotherapy can be self-applied easily by the patient at home (such as the use of heat packs), and may also be combined with other rehabilitation interventions.There is only 1 review paper of thermotherapy in the treatment of knee osteoarthritis nearly 10 years ago. Since only 3 articles with more than 20 years have been included, there have been reports in recent years that different forms of thermotherapy have been widely used in clinical practice. It has been used to treat knee osteoarthritis and has achieved significant results, especially a kind of thermotherapy called moxibustion, which is commonly used in Chinese hospitals.
Therefore, it is necessary to reassess the efficacy and safety of thermotherapy for KOA. In this study, evidence-based medicine will be used to analyze and evaluate clinical randomized controlled trials (RCTs) in patients with KOA.
# Methods
This protocol for this review was developed in accordance with the PRISMA-P guidelines and the Cochrane Handbook. This protocol has been registered on INPLASY (registration number: INPLASY202140038: https://inplasy.com/inplasy-2021-4-0038/ ). Ethical approval is unnecessary because this is a literaturebased study.
2.1. Inclusion criteria for study selection 2.1.1. Types of studies. All RCTs of thermotherapy for KOA without publication status restriction or writing language. Non-RCTs, quasi-RCTs, uncontrolled trials, reviews, case-controlled studies, animal trials, and laboratory studies will be excluded.
## Types of patients.
We will include participants with osteoarthritis of the knee (as defined by the study). There will also be no limitations related to age, sex, disease duration, and disease severity 2.1.3. Types of interventions. Interventions using thermotherapy only were included in this review. Trials that compared thermotherapy with standard treatment and/or placebo were included. thermotherapy with another active therapy vs the same therapy alone will also be investigated. Trials comparing head to head therapies, such as 2 different types of diathermy, were not included in this review.
## Western ontario and mcmaster universities osteoarthritis
Index is a self-report questionnaire for OA of the hip or knee, with higher scores indicating more serious pain, poorer physical function, and increased stiffness. It has been widely used as a tool by clinical investigators to assess patients with KOA.
## Secondary outcomes. lequesne index and medical
Outcomes Study Short Form 36 health survey will be accepted as the secondary outcomes.
2.1.4.3. Safety outcomes. The incidence and severity of side effects will be used to evaluate the safety. Any unexpected events occurred will be recorded. There will be no limitation to study publication status or language. The search terms include KOA, gonarthrosis, osteoarthrosis, osteoarthropathy, arthralgia, thermotherapy, diathermy, heat therapy, Moxibustion, and RCTs. The equivalent search words will be used in the Chinese databases. The detailed strategies for searching the PubMed database will be presented in .
## Search
## Searching other resources.
Additionally, the international clinical trials registry platform, dissertation, and gray literature will also be searched to identify systematic reviews related to thermotherapy for KOA. The relevant conference papers, journals will be retrieved manually.
2.3. Data collection and analysis 2.3.1. Selection of studies. For the first version of the review, 2 authors will screen the search yield and identify any potentially eligible citations. We will retrieve the full text of articles judged as being potentially eligible by at least 1 review author. Two review authors will screen the full-text articles for eligibility and resolve any disagreements by discussion or by consulting a third review author.We will record the selection process in sufficient detail to complete a study selection flow diagram.
## Data extraction and management.
Before data extraction, a standard form will be prepared for data collection. Two researchers will independently extract data of the included studies and write on the form. Any disagreement will be solved by consensus. The following data will be extracted: the first author, publication year, participants characteristics, interventions, duration of treatment, follow-up, outcome assessment, research results, adverse events, and other detail information. We will contact the original author for complete information when necessary.
## Assessment of risk of bias. assessment of risk of bias.
Two researchers will assess the risk of bias of included studies independently according to the Cochrane collaboration's tool.The tool comprise 7 aspects which are random sequence generation, allocation concealment, the blinding method for patients, researchers and outcomes assessors, incomplete outcome data, and selective reports.Every risk of bias will be classified as low, unclear, and high.
## Measures of treatment effect.
For continuous data, a mean difference or standardized mean difference with 95% confidence intervals (CIs) will be applied. For dichotomous outcome data, the risk ratio with 95% CIs will be used to evaluate the treatment effect. Search strategy used in PubMed.
## Search
Search terms If the essential data are not provided, we will try to contact the corresponding author of the articles by email for complete data. If the missing data cannot be obtained, we will analyze the available data.
2.3.6. Assessment of heterogeneity. We will assess clinical and methodological diversity in terms of participants, interventions, outcomes, and study characteristics for the included studies, to determine whether a meta-analysis is appropriate.When meta-analysis is appropriate, we will assess and quantify the possible magnitude of inconsistency (i.e., heterogeneity) across studies, using the I 2 statistic with a guide for interpretation as follows: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity; and 75% to 100% represents considerable heterogeneity. If we identify cases of considerable heterogeneity (defined as I 2 of 75% or greater), we will explore the data further by comparing the characteristics of individual studies and performing subgroup analyses.
2.3.7. Data synthesis. We plan to pool outcomes from trials with similar characteristics (participants, interventions and common comparators, outcome measures, and timing of outcome measurement) to provide estimates of benefit and harm. We plan to synthesize effect estimates using a randomeffects meta-analysis model based on the assumption that clinical diversity is likely to exist, and that different studies are estimating different intervention effects. Where we cannot pool data, we plan to present effect estimates and 95% CIs of each trial in tables, and summarize the results in text.
2.3.8. Sensitivity analysis. Sensitivity analysis. When there are sufficient studies, sensitivity analysis will be performed to assess the robustness of studies according to methodological quality, sample size, and missing data.
2.3.9. Reporting bias. In order to determine whether outcome reporting bias is present, we will check a priori trial protocols against published reports of trial results (i.e., check if all planned outcomes have results reported). We will compare the fixed-effect estimate against the random-effects model to assess the possible presence of small-sample bias in the published literature (i.e., in which the intervention effect is more beneficial in smaller studies).In the presence of small-sample bias, the randomeffects estimate of the intervention is more beneficial than the fixed-effect estimate. If we are able to pool more than 10 trials, we will undertake formal statistical tests to investigate funnel plot asymmetry to detect the possibility of publication bias.
2.3.10. Confidence in cumulative evidence. The quality of evidence will be assessed based on the grading of recommendations assessment, development, and evaluation system, include 4 levels: high, moderate, low, or very low.
## Interpreting results and reaching conclusions
We will follow the guidelines in Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions for interpreting results, and will be aware of distinguishing a lack of evidence of eHect from a lack of eHect.We will base our conclusions only on findings from the quantitative or narrative synthesis of included studies for this review. We will avoid making recommendations for practice, and our implications for research will suggest priorities for future research and outline what the remaining uncertainties are in the area.
# Author contributions
Conceptualization: Na Li, Jinsen Xu. Data curation: Na Li, Bin Chen. Formal analysis: Zhining Wu, Jinsen Xu. Funding acquisition: Jinzuan Wu. |
Reduction of environmental pollutants for prevention of cardiovascular disease: it’s time to act
# Introduction
Cardiovascular disease (CVD) represents the result of underlying genetic predisposition and lifetime exposure to multiple environmental factors. The past century has seen a revolution in our understanding of the importance of modifiable risk factors such as diet, exercise, and smoking. Exposure to environmental pollutants, be it in the air, water, or physical environment, is increasingly recognized as a silent, yet important determinant of CVD.The quote 'genetics loads the gun but the environment pulls the trigger', put forward by G.A. Bray and F. Collins, exemplifies the complex relationship between human disease and the environment. The cardiovascular system is highly vulnerable to a variety of environmental insults, including tobacco smoke, solvents, pesticides, and other inhaled or ingested pollutants, as well as extremes in noise and temperature. While our understanding of multiple environmental factors continues to evolve, it is estimated that environmental air pollution and noise pollution alone may contribute to a substantial burden attributable to environmental factors as we currently understand them. It is important to note that noise and air pollution can have many of the same sources such as heavy industry, road and aircraft vehicles. In a recent in-depth report, the European Commission acknowledged that the societal costs for the combination noise and air pollution are nearly 1 trillion Euros, while the costs for alcohol and smoking are considerably less (50-120 and 540 billion Euro, respectively, see https://ec.europa. eu/environment/integration/research/newsalert/pdf/air_noise_pol lution_socioeconomic_status_links_IR13_en.pdf).
The World Health Organization (WHO) calculates that 12.6 million premature deaths per year are attributable to unhealthy environments, 8.2 million of which are due to non-communicable disease, with CVD (including stroke) being the largest contributor, accounting for nearly 5 million of these deaths. [bib_ref] Drug treatment of elderly patients with acute myocardial infarction: practical recommendations, Aronow [/bib_ref] Among all environmental pollutants, poor air quality is the most important risk factor, and ambient air pollution due to particulate matter <2.5 mm (PM 2.5 ) exposure ranks 5th among all global risk factors in 2015, leading to 4.2 million deaths annually as estimated by the Global Burden of Disease study. [bib_ref] Estimates and 25-year trends of the global burden of disease attributable to..., Cohen [/bib_ref] Nine out of 10 people worldwide are exposed to ambient air pollutant levels above WHO guidelines (>10 mg/m). [bib_ref] Estimates and 25-year trends of the global burden of disease attributable to..., Cohen [/bib_ref] [bib_ref] KF31327, a new potent and selective inhibitor of cyclic nucleotide phosphodiesterase 5, Hirose [/bib_ref] Using a novel exposure-response hazard function (global estimate of exposure mortality model) to estimate global mortality attributable to air pollution, Burnett et al. [bib_ref] Global estimates of mortality associated with long-term exposure to outdoor fine particulate..., Burnett [/bib_ref] and Lelieveld et al. [bib_ref] Loss of life expectancy from air pollution compared to other risk factors:..., Lelieveld [/bib_ref] found that around 9 million global premature deaths (790 000 excess deaths in Europe alone) were attributable to air pollution, [bib_ref] Air pollution, chronic smoking, and mortality, Lelieveld [/bib_ref] numbers that are well comparable to that of smoking. [bib_ref] Loss of life expectancy from air pollution compared to other risk factors:..., Lelieveld [/bib_ref] These figures are substantially higher than those estimated by the WHO and Global Burden of Disease study. [bib_ref] Drug treatment of elderly patients with acute myocardial infarction: practical recommendations, Aronow [/bib_ref] [bib_ref] Estimates and 25-year trends of the global burden of disease attributable to..., Cohen [/bib_ref] Ambient noise is the other omnipresent exposure with emerging data suggesting a large attributable burden of disability to this factor in many urban environments. In Western Europe, it is estimated that around 1.6 million healthy life years are lost every year due to noise. It is estimated that a large part of the European population is exposed to noise originating from road traffic at levels exceeding 55 decibels [dB(A), A-weighted decibel scale adapted to the human hearing frequencies]; 20% exposed to levels exceeding 65 dB(A) during the daytime; and 30% of the population is exposed to levels exceeding 55 dB(A) (see https://www.eea.europa.eu/publications/environmen tal-noise-in-europe). In this review, we will focus on the cardiovascular effects of ambient air pollution and noise pollution as prototypical environmental factors that provide important lessons to facilitate understanding of the outsize effects of the environment on susceptibility to CVD. The pathophysiology, epidemiology, mitigation measures, and future challenges for these two common yet pervasive environmental factors are discussed in detail.
In many parts of the world, a substantial portion of the urban population is exposed to road traffic noise at levels exceeding 55 dB . [bib_ref] Are air pollution and traffic noise independently associated with atherosclerosis: the Heinz..., Kalsch [/bib_ref] In cities in Asia, the proportion of the population reaching L den levels (day-evening-night level, i.e. the average sound pressure level measured over a 24 h period with adjustment for more detrimental health effects of nocturnal noise) of 60-64 dB is very high. [bib_ref] Quantification of the exposure and effects of road traffic noise in a..., Brown [/bib_ref] In contrast to the relatively straightforward classification of noise, air pollution is intrinsically complex and defy easy classification. From a regulatory perspective, 'criteria' air pollutants allow health-based and/or environmentally based guidelines for setting permissible levels.These include carbon monoxide, lead, nitrogen oxides, groundlevel ozone, particle pollution (often referred to as PM), and sulphur oxides. Particulate matter is categorized based on its aerodynamic diameter: < _10 lm [thoracic particles (PM 10 )], < _2.5 lm [fine particles (PM 2.5 )], < _0.1 lm [ultrafine particles (UFP)], and between 2.5 and 10 lm [coarse particles (PM 2.5-10 )]. Although 'criteria' pollutants are regulated individually, it is anticipated that the effects of air pollution are driven by the complex interaction of particulate and gaseous components in mixtures and that smaller particles (e.g. UFP) are more detrimental then larger ones.
There is substantial spatial and temporal variation of both noise and air pollution. Traffic-related pollutants and noise often peaking during the late morning and evening rush hours. Gradients for both noise and air pollutants are also dependent upon meteorological conditions, including diurnal changes in vertical mixing height, wind speed, and temperature. In the case of noise, the gradients are substantial as the intensity of noise decreases exponentially with the distance from its source. The gradients for air pollution from their source may also differ depending upon the pollutant. Traffic factors, such as the speed, traffic load, etc., may also differentially affect noise and traffic-related air pollution. During traffic congestion, when traffic is at standstill or at lower engine speeds, noise levels may be lower, but emissions may be dramatically higher, contributing to marked surges in trafficrelated air pollutants. In contrast, when traffic is moving well, noise levels may be higher, but emissions may be lower. Environmental factors such as road conditions, noise barriers, and surrounding buildings are well known to influence traffic noise but may not influence air pollution substantially.
The highly associated nature of traffic noise and air pollution makes it challenging to isolate their independent effects on cardiovascular events in epidemiological studies. A few studies have attempted to assess the independent contribution of noise from air pollution and vice versa. The results are, however, somewhat variable, with some studies demonstrating an independent effect of noise and/or air pollution on cardiovascular morbidity and mortality, while others find marked attenuation of effects after adjusting for the other. Whether noise and air pollution have differing, additive, synergistic, and/or confounding effects upon cardiovascular health is still incompletely understood. Also of great importance in all air pollution and noise exposure studies is the co-linearity of these risk factors to other confounders (e.g. lower socio-economic status, psychosocial stressors, other poorly understood environmental variables and adverse lifestyle factors) that often go handin-hand with pollutants.
Pathophysiology and epidemiology of noise and cardiovascular disease Epidemiology During the last decade, a number of epidemiological studies have investigated effects of transportation noise on risk for CVD. In 2018, a systematic review by WHO found that there was substantial evidence to conclude that road traffic noise increases the risk for ischaemic heart disease, with an 8% higher risk per 10 dB higher noise. [bib_ref] WHO environmental noise guidelines for the European region: a systematic review on..., Kempen [/bib_ref] For stroke, the evidence was ranked as moderate, with only one study on incidence and four on mortality. [bib_ref] WHO environmental noise guidelines for the European region: a systematic review on..., Kempen [/bib_ref] Subsequently, large population-based studies from Frankfurt, London, and Switzerland found road traffic noise to increase stroke incidence and/or mortality, especially ischaemic strokes, [bib_ref] The effect of aircraft, road, and railway traffic noise on stroke-results of..., Seidler [/bib_ref] [bib_ref] Road traffic noise is associated with increased cardiovascular morbidity and mortality and..., Halonen [/bib_ref] [bib_ref] Probst-Hensch N, Röö sli M; SNC Study Group. Transportation noise exposure and..., Héritier [/bib_ref] whereas smaller cohort studies indicated no association. [bib_ref] Road traffic noise, air pollution and incident cardiovascular disease: a joint analysis..., Cai [/bib_ref] Recently, road traffic noise has been found to increase the risk for other major CVD not evaluated by WHO, most importantly heart failure and atrial fibrillation. [bib_ref] Probst-Hensch N, Röö sli M; SNC Study Group. Transportation noise exposure and..., Héritier [/bib_ref] [bib_ref] Residential exposure to traffic noise and risk of incident atrial fibrillation: a..., Monrad [/bib_ref] Aircraft noise has also been associated with higher CVD incidence and mortality, 14,17 but due to a limited number of studies, the evidence is still rated low to moderate. [bib_ref] WHO environmental noise guidelines for the European region: a systematic review on..., Kempen [/bib_ref] Epidemiological studies have linked transportation noise with a number of major cardiovascular risk factors, most consistently obesity and diabetes. [bib_ref] Association between noise exposure and diabetes: a systematic review and meta-analysis, Sakhvidi [/bib_ref] [bib_ref] Long-term exposure to transportation noise in relation to development of obesity-a cohort..., Pyko [/bib_ref] Also, many studies investigated effects of noise on hypertension, and although a meta-analysis of 26 studies found that road traffic noise was associated with higher prevalence of hypertension, 11 studies on incidence are still few and inconsistent.
Ambient air pollution and traffic noise, especially from roads, are correlated and suspected of being associated with the same CVD, and therefore mutual adjustment is highly important. Most recent studies on noise and CVD adjust for air pollution and generally the results are found to be robust to the adjustment, suggesting that transportation noise is indeed an independent risk factor for CVD. [bib_ref] Long-term residential road traffic noise and mortality in a Danish cohort, Thacher [/bib_ref] Another noise source investigated in relation to CVD risk is occupational noise; an exposure mainly occurring during daytime. Most existing studies are cross-sectional, and results from a few prospective studies providing conflicting evidence, with some studies indicating an association with CVD, [bib_ref] Longitudinal study of occupational noise exposure and joint effects with job strain..., Eriksson [/bib_ref] whereas others finding no association, [bib_ref] Occupational noise exposure and the risk of stroke, Stokholm [/bib_ref] stressing the need for more well-designed prospective studies.
## Pathophysiology
According to the noise stress reaction model introduced by Babisch, 24 non-auditory health effects of noise have been demonstrated to activate a so-called 'indirect pathway', which in turn represents the cognitive perception of the sound, and its subsequent cortical activation is related to emotional responses such as annoyance and anger (reviewed in Ref. [bib_ref] Environmental stressors and cardio-metabolic disease: part IImechanistic insights, Munzel [/bib_ref] This stress reaction chain can initiate physiological stress responses, involving the hypothalamus, the limbic system, and the autonomic nervous system with activation of the hypothalamus-pituitary-adrenal (HPA) axis and the sympathetic-adrenal-medulla axis, and is associated with an increase in heart rate and in levels of stress hormones (cortisol, adrenalin, and noradrenaline) enhanced platelet reactivity, vascular inflammation, and oxidative stress (see . While the conscious experience with noise might be the primary source of stress reactions during daytime (for transportation and occupational noise), the subconscious biological response during night-time in sleeping subjects, at much lower transportation noise levels, is thought to play an important role in pathophysiology, particularly through disruption of sleep-wake cycle, diurnal variation, and perturbation of time periods critical for physiological and mental restoration. Recent human data provided a molecular proof of the important pathophysiological role of this 'indirect pathway' by identifying amygdalar activation (using [bib_ref] WHO environmental noise guidelines for the European region: a systematic review on..., Kempen [/bib_ref] F-FDGPET/CT imaging) by transportation noise in 498 subjects, and its association with arterial inflammation and major adverse cardiovascular events.These data are indeed consistent with animal experiments demonstrating an increased release of stress hormones (catecholamines and cortisol), higher blood pressure, endothelial dysfunction, [bib_ref] Krö ller-Schön S. Effects of noise on vascular function, oxidative stress, and..., Münzel [/bib_ref] neuroinflammation, diminished neuronal nitric oxide synthase (nNOS) expression as well as cerebral oxidative stress in aircraft noise-exposed mice. [bib_ref] Crucial role for Nox2 and sleep deprivation in aircraft noise-induced vascular and..., Krö Ller-Schö N [/bib_ref] These changes were substantially more pronounced when noise exposure was applied during the sleep phase (reflecting night-time noise exposure) and was mostly prevented in mice with genetic deletion or pharmacological inhibition of the phagocytic NADPH oxidase (NOX-2). [bib_ref] Crucial role for Nox2 and sleep deprivation in aircraft noise-induced vascular and..., Krö Ller-Schö N [/bib_ref] These studies also revealed substantial changes in the gene regulatory network by noise exposure, especially within inflammatory, antioxidant defence, and circadian clock pathways . [bib_ref] Krö ller-Schön S. Effects of noise on vascular function, oxidative stress, and..., Münzel [/bib_ref] [bib_ref] Crucial role for Nox2 and sleep deprivation in aircraft noise-induced vascular and..., Krö Ller-Schö N [/bib_ref] The conclusions from these experiments are supportive of a role for shortened sleep duration and sleep fragmentation in cerebrovascular oxidative stress and endothelial dysfunction.
## Figure 1
The key mechanisms of the adverse health effects of traffic noise exposure. Environmental noise exposure causes mental stress responses, a neuroinflammatory phenotype, and cognitive decline. This may lead to manifest psychological disorders and mental diseases or, via stress hormone release and induction of potent vasoconstrictors, to vascular dysfunction and damage. All of these mechanisms initiate cardio-metabolic risk factors that lead to manifest end organ damage. Of note, chronic cardio-metabolic diseases often are associated with psychological diseases and vice versa. [bib_ref] Environmental noise-induced effects on stress hormones, oxidative stress, and vascular dysfunction: key..., Hahad [/bib_ref] - ACTH, adrenocorticotropic hormone; ADH, antidiuretic hormone (vasopressin); ATII, angiotensin II; CRH, corticotropin-releasing hormone; eNOS, endothelial nitric oxide synthase; ET-1, endothelin-1;NO, nitric oxide; NOX-2, phagocytic NADPH oxidase (catalytic subunit).
## Current opinion reduction of environmental pollutants
Likewise, we observed a significant degree of endothelial dysfunction, an increase in stress hormone release, blood pressure and a decrease in sleep quality in healthy subjects and patients with established coronary artery disease, in response to night-time aircraft noise (reviewed in Ref. [bib_ref] Environmental stressors and cardio-metabolic disease: part IImechanistic insights, Munzel [/bib_ref] Importantly, endothelial dysfunction was corrected by the antioxidant vitamin C indicating increased vascular oxidative stress in response to night-time aircraft noise exposure. The important role of oxidative stress and inflammation for noiseinduced cardiovascular complications was also supported by changes of the plasma proteome, centred on redox, pro-thrombotic and proinflammatory pathways, in subjects exposed to train noise for one night [mean SPL 54 dB ]. [bib_ref] Acute exposure to nocturnal train noise induces endothelial dysfunction and prothromboinflammatory changes..., Herzog [/bib_ref] Pathophysiology and epidemiology of air pollution and cardiovascular disease
Since the publication of an American Heart Association Scientific Statement, [bib_ref] American Heart Association Council on Epidemiology and Prevention, Council on the Kidney..., Brook [/bib_ref] there has been a consistent stream of epidemiological and mechanistic evidence linking PM 2.5 , the most frequently implicated air pollution component with CVD. [bib_ref] Global estimates of mortality associated with long-term exposure to outdoor fine particulate..., Burnett [/bib_ref] [bib_ref] Loss of life expectancy from air pollution compared to other risk factors:..., Lelieveld [/bib_ref] Mounting evidence suggests that health risks attributable to PM 2.5 persist even at low levels, below WHO air quality guidelines and European standards (annual levels <10 and <25 mg/m 3 , respectively). Updated exposureresponse dose curves suggest a robust supralinear concentrationresponse-curve for PM and CVD with no apparent safe threshold level. [bib_ref] Environmental determinants of cardiovascular disease: lessons learned from air pollution, Al-Kindi [/bib_ref] Epidemiology Current estimates suggest air pollution is associated with around 9 million premature deaths, worldwide annually with 40-60% of mortality attributed to cardiovascular causes. 5,33 Short-term exposure (over hours or days) is associated with increased risk for myocardial infarction, stroke, heart failure, arrhythmia, and sudden death by about 1-2% per 10 mg/m 3 . Longer-term exposure over months or years, amplifies these risk associations, to 5-10% per 10 mg/m 3 . Living in regions with poor air quality potentiates the atherosclerotic process and promotes the development of several chronic cardiometabolic conditions (e.g. diabetes, hypertension).
Although the strength of the association for criteria air pollutants is strongest for PM 2.5 , there are data linking other pollutants such as nitrogen oxides (e.g. NO 2 ) and less consistently ozone (O 3 ) with cardiovascular events. [bib_ref] Environmental determinants of cardiovascular disease: lessons learned from air pollution, Al-Kindi [/bib_ref] Pollutants from traffic and combustion sources are of high concern (due to high levels of ultrafine PM, toxicity of constituents, and penetration of pollutants systemically) although precise burden estimates have yet to be established for this source. Coarse PM 10 air pollution from anthropogenic sources has been associated with cardiovascular disease although sources such as agricultural emissions and crustal material are less well studied.
Given the continuing links between PM 2.5 and adverse cardiovascular events, even at levels substantially below 10 mg/m 3 , there is a need for a realistic lower limit that may strike the balance between what is reasonably possible and eliminating anthropogenic sources. It is important to keep in mind that complete elimination of all PM2.5 may not possible given that some PM 2.5 is natural. Calculations by Lelieveld et al. [bib_ref] Cardiovascular disease burden from ambient air pollution in Europe reassessed using novel..., Lelieveld [/bib_ref] of a complete phase-out of fossil fuel-related emissions (needed to achieve the 2 C climate change goal under the Paris Agreement) demonstrated a reduction in excess mortality rate of 3.61 million per year worldwide. The increase in mean life expectancy in Europe would be around 1.2 years indicating a tremendous health co-benefit from the phase-out of carbon dioxide emissions.
## Pathophysiology
Mechanistic studies, using controlled exposure studies in humans and experimental models support a causal relationship between PM and CVD. Acute exposure to air pollutants induces rapid changes that include vasoconstriction, endothelial dysfunction, arterial stiffening, arrhythmia, exacerbation of cardiac ischaemia, increased blood coagulability, and decreased fibrinolytic capacity. Additionally, longterm exposure to PM accelerates the growth and vulnerability of atherosclerotic plaques. [bib_ref] Air pollution and cardiovascular disease: car sick, Miller [/bib_ref] A broad range of mechanisms accounts for pathophysiology at an organ and cellular level, with inflammation and oxidative stress playing key roles. [bib_ref] Environmental stressors and cardio-metabolic disease: part IImechanistic insights, Munzel [/bib_ref] Additionally, several convincing pathways can account for the link between inhalation of pollutants and the cardiovascular system, including passage of inflammatory (and other) mediators into the circulation, direct passage of particles (or their constituents) into circulation, imbalance of autonomic nervous system activity, and changes to central control of endocrine systems. The contribution of individual pathways will depend on type of pollutant, the exposure (dose and duration), specific cardiovascular endpoints, and the health status of individual. Finally, the cardiovascular effects of pollutants occur in both healthy individuals and those with pre-existing cardiorespiratory disease, suggesting a potential contributory role on the induction, progression, and exacerbation of CVD. [bib_ref] Environmental determinants of cardiovascular disease: lessons learned from air pollution, Al-Kindi [/bib_ref] [bib_ref] Air pollution and cardiovascular disease: car sick, Miller [/bib_ref] Mitigation strategies
## Noise mitigation
In 2020, the European Environment Agency concluded that more than 20% of the EU population live with road traffic noise levels that are harmful to health and that this proportion is likely to increase in the future (see https://www.eea.europa.eu/publications/environmen tal-noise-in-europe [last accessed 17/09/2020]). European Environment Agency also estimated that in EU, 22 million live with high railway noise and 4 million with high aircraft noise.
The authorities can use different strategies to reduce levels of traffic noise [fig_ref] Table 1: Mitigation methods resulting in reduction in road traffic noise [/fig_ref]. For road traffic, the sound generated by the contact between the tires and the pavement is the dominant noise source, at speeds above 35 km/h for cars and above 60 km/h for trucks. Therefore, changing to electric cars will result in only minor reductions in road traffic noise. Generally applied strategies for reducing road traffic noise include noise barriers in densely populated areas, applying quiet road surfaces, and reducing speed, especially during night-time. Furthermore, there is a great potential in developing and using low-noise tires. As many of these mitigation methods result in only relatively small changes in noise [fig_ref] Table 1: Mitigation methods resulting in reduction in road traffic noise [/fig_ref] , a combination of different methods is important in highly exposed areas. traffic routes and housing zones, introduction of night bans, and implementation of continuous descent arrivals, which require the aircraft to approach on steeper descents with lower, less variable throttle settings. For railway noise, replacing cast-iron block breaks with composite material, grinding of railway tracks and night bans, are among the preferred strategies for reducing noise. Lastly, installing sound-reducing windows and/or orientation of the bedroom towards the quiet side of the residence can reduce noise exposure.
## Air pollution mitigation
Although it is widely recognized that legislation, policies, regulation, and technology, coupled with enforcement, are critical to reduction of air pollution levels, the political momentum required to accomplish this globally is currently limited. Thus, personal measures to mitigate risk take on a much greater importance. The current experience and lessons learned with personal protective equipment and mitigation in reducing exposure to SARS-CoV2 are highly reminiscent of their use in combating air pollution, albeit the protection provided varies depending on the pollutant. [bib_ref] Flattening the curve in COVID-19 using personalised protective equipment: lessons from air..., Rajagopalan [/bib_ref] Mitigation measures must be affordable and broadly applicable to the population, and the level of protection provided should match the risk of population that is being exposed [fig_ref] Figure 2: Mitigation measures to reduce air pollution exposure [/fig_ref]. The latter would necessitate an understanding of the health risk of the patient/community and degree of exposure. The need and urgency plus intensity of any recommended intervention also need to be weighed against their potential benefits vs. risks for each individual (e.g. wasted effort, resources, unnecessary concern, or possible complacency of the user). Although no intervention to reduce air pollution exposure has as yet been shown to reduce cardiovascular events, the consistent link between increased levels of PM 2.5 and cardiovascular events, evidence for measures in lowering PM 2.5 levels, and the impact of several mitigation strategies in improving surrogate markers are highly suggestive that interventions could be correspondingly impactful in reducing cardiovascular events.
Current approaches to mitigate air pollution and their impact have been previously reviewed and can be broadly classified into: (i) Active personal exposure mitigation with home air cleaning and personal equipment; (ii) Modification of human behaviour to reduce passive exposures; (iii) Pharmacologic approaches. [bib_ref] Environmental determinants of cardiovascular disease: lessons learned from air pollution, Al-Kindi [/bib_ref] Studies on N95 respirator under ambient PM 2.5 exposure conditions at both high and low levels of exposures over a few hours have shown to reduce systolic blood pressure and improve heart rate variability. [bib_ref] Environmental determinants of cardiovascular disease: lessons learned from air pollution, Al-Kindi [/bib_ref] [bib_ref] Reducing personal exposure to particulate air pollution improves cardiovascular health in patients..., Langrish [/bib_ref] In the only trial comparing exposure mitigation to both noise and air pollution, individual reduction of air pollution or noise with a respirator or noise-cancelling headphones, respectively, did not alter blood pressure. Heart rate variability indices were, however, variably improved with either intervention. [bib_ref] Cardiovascular benefits of reducing personal exposure to traffic-related noise and particulate air..., Yang [/bib_ref] Face masks and procedural masks (e.g. surgical masks) are widely available but are not effective in filtering PM 2.5 , especially if poorly fitting or worn during high activity, [bib_ref] Effectiveness of face masks used to protect Beijing residents against particulate air..., Cherrie [/bib_ref] and therefore cannot be recommended for widespread usage if N95 respirators are available. Closing car windows, air-conditioning, and cabin air filters represent approaches that could be important in those who are susceptible, but only in those spending large amounts of time in transportation microenvironments. Behavioural strategies such as air pollution avoidance by changing travel routes, staying indoors/closing windows, and modification of activity can help limit air pollution exposure, but unintended consequences in some instances have the potential of offsetting benefit. An example is closing windows to limit outdoor exposure but increasing the hazard for indoor air pollutants or limiting outdoor recreation/exercise to mitigate ambient exposures. The latter scenario of limiting outdoor exposure brings up some very practical questions about the risk/benefit of loss of cardiovascular benefits of exercise vs. potential gain from benefits secondary to air pollution mitigation. Health impact modelling and epidemiologic studies have demonstrated that the benefits of aerobic exercise nearly always exceed the risk of air pollution exposure across a range of concentrations, and for long durations of exercise for normal individuals (>75 min). Based on current evidence, guiding healthy people to avoid outdoor activity in areas with high PM 2.5 pollution has the potential to produce greater harm than benefit, given the low absolute risk for cardiovascular or respiratory events. On the other hand, advising patients with pre-established CVD to continue to remain >400 m away from major roadways to avoid exposure to traffic pollutants is a reasonable measure, despite the current lack of strong evidentiary support. Although a variety of over the counter drugs and medications have been shown to mitigate association between air pollution and surrogates, almost none can be recommended to protect against air pollution mediated adverse health effects at this time. However, the use of medications for primary and secondary prevention of CHD should be encouraged if indicated for other reasons.
Housing and urban design to improve cardiovascular health Two-third of the European population live in urban areas and this number continues to grow. A recent Statement on Air Quality Policy has discussed aspects in the built environment that may be targeted in order to reduce exposures to PM 2.5 (in press 2020). Briefly, built environment features may directly or indirectly modify adverse cardiovascular effects of air pollution through the indoor living environment, green spaces, roads, utilities, and transportation infrastructure. The design of communities has the potential of impacting exposures, by affecting the continuum of human existence across indoor living, commuting, working, and recreation [fig_ref] Figure 3: Urban design considerations to reduce exposure to noise and air pollution [/fig_ref]. The layout of roads, sidewalks, green spaces, and the availability of cheap public transportation can affect travel behaviour and can help alleviate air quality.Communities with proximity and compactness have been associated with higher life expectancy, improved air quality, and health. [bib_ref] Associations between Urban Sprawl and Life Expectancy in the United States, Hamidi [/bib_ref] [bib_ref] Urban form, air pollution, and health, Hankey [/bib_ref] Green environments can improve air quality, encourage physical activity, and promote social interactions, ultimately improving cardiovascular health. Indeed, there is evidence to support a protective association of green spaces on PM-associated CVD. [bib_ref] The influence of green space on the short-term effects of particulate matter..., Heo [/bib_ref] [bib_ref] Neighborhood greenness attenuates the adverse effect of PM2.5 on cardiovascular mortality in..., Yitshak-Sade [/bib_ref] All-cause and ischaemic heart disease mortality related to income deprivation has been shown to be lower in populations who live in the greenest areas, vs. those who have less exposure to green space. [bib_ref] Effect of exposure to natural environment on health inequalities: an observational population..., Mitchell [/bib_ref] Recently, Giles-Corti identified eight integrated regional and local interventions that, when combined, encourage walking, cycling and public transport use, while reducing private motor vehicle use. [bib_ref] City planning and population health: a global challenge, Giles-Corti [/bib_ref] These eight interventions are directed to reduce traffic exposure, to reduce air pollution and noise, and to reduce the important public health issue loneliness and social isolation, to improve the safety from crime, to reduce physical inactivity and prolonged sitting, and to prevent the consumption of unhealthy diets. 45
## Future perspectives: opportunities and challenges over the next decade
Efforts to mitigate air pollution and noise are endeavours that involve complex economic and geopolitical considerations. Measures such as transportation reform, shift to zero-emission fuels, urban landscape reform, and ecologically sound lifestyle changes may help simultaneously alleviate air/noise pollution while accomplishing climate change goals. However, reducing air pollution and noise may have short-term challenges due to economic incentives that are substantially misaligned with health and environmental priorities and thus opportunities to understand the importance of these factors in human health will sadly continue. An important avenue of investigation is convergent studies that look at the broad and collective impact and burden of air and noise pollution as archetypal environmental risk factors. The questions that need to be addressed are many and include the magnitude and time course of response of co-exposure, interactive effects of environmental factors on surrogate measures, duration of effect/time course of reversal, impact on circadian rhythm, and finally the effect of reversal as well as prevention and lifestyle approaches that may help mitigate risk (e.g. diet, stress, and exercise).
The rapid development of personalized technologies that provide multiple measures of health in fine temporal detail in conjunction with data on environmental exposure provide an unprecedented opportunity for research and may allow an extraordinary understanding of the interactions between environmental and non-environmental risk factors over long durations. Together with developments in next-generation sequencing technologies, and opportunities in big data, assimilative studies of this nature may finally provide a granular view of the environmental-genetic interactions leading to the development of CVD. However, the extent of these advances may be tempered by the need to manage subject burden and costs, and imprecise data on many environmental variables. Increased awareness of the societal burden posed by environmental risk factors and acknowledgement in traditional risk factor guidelines may pressurize politicians to intensify the efforts required for effective legislation.
The cardiovascular community has a responsibility to help promulgate the impact of, not only health lifestyle and diet, but also over the outsize impact of air and noise pollution on cardiovascular health. Individuals can apply political pressure through democratic means Take home figure Upper left panel reproduced from Münzel et al. [bib_ref] Environmental factors such as Noise and Air Pollution and Vascular Disease, Münzel [/bib_ref] with permission. and lobbying to enact changes at regional and national levels that lead to reductions in noise/air pollution exposure. Patient organization can provide a strong voice in the call for action at governmental level. Importantly, air pollution was mentioned in the published guidelines for cardiovascular prevention, but the recommendations to reduce pollution were completely insufficient,while prevention measures with respect to traffic noise were completely lacking. Noise and air pollution represent significant cardiovascular risk factors, it is important that these factors are included into the ESC guidelines, and others, for myocardial infarction, arterial hypertension, and heart failure.
[fig] Figure 2: Mitigation measures to reduce air pollution exposure. [/fig]
[fig] Figure 3: Urban design considerations to reduce exposure to noise and air pollution. Current opinion reduction of environmental pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [/fig]
[table] Table 1: Mitigation methods resulting in reduction in road traffic noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [/table]
[table] Table 2: Personal active mitigation methods to reduce air pollution exposure Type of intervention Efficacy in reducing exposure Considerations for use Evidence in reducing surrogate outcomes Personal air purifying respirators (reducing solid but not gaseous air pollutants). Fit and use frequency are key determinants of efficacy. A valve or microventila-Designed to clean air in a small area. Effective in reducing indoor particles but duration of use and volume of room, key determinants of efficacy. Effective in reducing concentrations as long as filters replaced regularly. Efficacy is variable with building and operational factors (i.e. open windows) No data currently available [/table]
|
Critical Analysis of the Current Medical Image-Based Processing Techniques for Automatic Disease Evaluation: Systematic Literature Review
# Introduction
Classification methods have increased in importance and now play a significant role in image processing. Their importance stems from their applications in various fields, particularly in medicine. Given the importance of classification in medicine, new and sophisticated classification tools and methods are needed to diagnose and classify medical images efficiently [bib_ref] Performance research on medical data classification using traditional and soft computing techniques, Ansari [/bib_ref]. Several classification algorithms encompass hundreds of different classification issues, and no single classification method can successfully and efficiently address all classification problems. As a result, answering the question concerning which classification approach is best for a particular study is challenging. The fast growth in medical data and imagery in recent years has necessitated the employment of new methodologies depending on big data technology, artificial intelligence, and machine learning in health care, making it an important research area [bib_ref] Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning, Kermany [/bib_ref]. Given the importance of classification in the medical field, new approaches for rapidly identifying and evaluating medical images are required. As a result, this research aims to compare existing and conventional methods for medical image classification and, based on these findings, suggest a novel algorithm for medical image classification [bib_ref] Medical image analysis using deep learning: A systematic literature review, Sudheer Kumar [/bib_ref].
The field of medical image processing and analysis has contributed to substantial medical achievements. A correct diagnosis necessitates the precise identification of each disease by integrating methods and techniques that support more effective clinical diagnosis depending on images obtained by various imaging modalities that have been used increasingly widely and successfully to detect illnesses [bib_ref] Deep learning-based breast cancer classification through medical imaging modalities: State of the..., Murtaza [/bib_ref]. This study aims to describe the process of medical image analysis, identify the techniques used in the analysis, and
## Research questions
The SLR (Systematic Literature Review) aims to address the research questions by finding all relevant research outcomes from previous studies. The research questions are divided into five sub-questions: Q1: What are the modalities of medical imaging? Q2: What is the task of medical image processing and analysis? Q3: Which medical image processing methods are most used in diagnostic systems? Q4: What diagnostic techniques have been adopted and developed? Q5: Is the system that has been adopted or designed capable of producing good results?
We searched various databases such as Elsevier, IEEE Explorer, Springer, and Google Scholar. We included the relevant studies that mainly focus on two or more questions based on our research questions.
## Research strategy
Our systematic literature review collected many studies related to our search topic over the last five years, between 2017 and 2021, from the following databases: Elsevier, IEEE Xplorer, Springer, and Google Scholar. We used combinations of keywords and terms, including "Medical Image Analysis", "Medical Image Processing", "Medical Image Processing Techniques", "Medical Image Processing Techniques" AND "Disease Diagnosis", and "Diagnostic Techniques" AND "Medical Image Processing", and we obtained about 3204 articles meeting our searched keywords. After removing 206 duplicate articles, the remaining set consisted of 2998 articles. After examining these studies by their titles and abstracts, a new set of keywords were applied, including "Diseases Classification", "Machine Learning", "Deep Learning", "Neural Networks", and "Hybrid Diagnosis System", and 2900 articles were excluded. Of the 98 articles remaining, 58 were excluded after deep reading and based on the exclusion criteria focused on articles that used machine learning and deep learning methods to create a hybrid diagnostic system based on the merging of two methods or the modification or improvement of a common method and its development for proposing possible areas for future research. These articles reached the fourth stage of our systematic exclusion technique and dealt with different topics and methods related to machine learning (ML) methods found in , deep learning (DL) strategies found in , and convolutional neural network (CNN) approaches found in [bib_ref] Boosting Breast Cancer Detection Using Convolutional Neural Network, Alanazi [/bib_ref] [bib_ref] Brain Tumor Classification Using Convolution Neural Network, Saranya [/bib_ref] [bib_ref] Classification of Brain Tumors from MRI Images Using a, Badža [/bib_ref] [bib_ref] Convolutional neural network for the diagnosis of early gastric cancer based on..., Li [/bib_ref] [bib_ref] A deep convolutional neural network based computer aided diagnosis system for the..., Sathiyamoorthi [/bib_ref] [bib_ref] Deep learning convolutional neural network (CNN) With Gaussian mixture model for predicting..., Sekaran [/bib_ref] [bib_ref] Skin cancer classification using Convolutional neural networks, Subramanian [/bib_ref]. Finally, 40 studies fulfilling our research criteria were obtained to be deeply analyzed. In this way, the most suitable articles were selected based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) technique, as shown in [fig_ref] Figure 1: Flowchart of the process for article selection [/fig_ref]. By applying the PRISMA technique, which is appropriate for any systematic literature review, we kept only the most relevant articles from large databases. In the final step, as shown in [fig_ref] Figure 1: Flowchart of the process for article selection [/fig_ref] , the last article set was not only a result of the automatic selection based on keyword combinations but also represented the answers to our research questions that are discussed in Section 2.3.
Sensors 2022, 3 of 25 methods related to machine learning (ML) methods found in , deep learning (DL) strategies found in , and convolutional neural network (CNN) approaches found in [bib_ref] Boosting Breast Cancer Detection Using Convolutional Neural Network, Alanazi [/bib_ref] [bib_ref] Brain Tumor Classification Using Convolution Neural Network, Saranya [/bib_ref] [bib_ref] Classification of Brain Tumors from MRI Images Using a, Badža [/bib_ref] [bib_ref] Convolutional neural network for the diagnosis of early gastric cancer based on..., Li [/bib_ref] [bib_ref] A deep convolutional neural network based computer aided diagnosis system for the..., Sathiyamoorthi [/bib_ref] [bib_ref] Deep learning convolutional neural network (CNN) With Gaussian mixture model for predicting..., Sekaran [/bib_ref] [bib_ref] Skin cancer classification using Convolutional neural networks, Subramanian [/bib_ref]. Finally, 40 studies fulfilling our research criteria were obtained to be deeply analyzed. In this way, the most suitable articles were selected based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) technique, as shown in [fig_ref] Figure 1: Flowchart of the process for article selection [/fig_ref]. By applying the PRISMA technique, which is appropriate for any systematic literature review, we kept only the most relevant articles from large databases. In the final step, as shown in [fig_ref] Figure 1: Flowchart of the process for article selection [/fig_ref] , the last article set was not only a result of the automatic selection based on keyword combinations but also represented the answers to our research questions that are discussed in Section 2.3.
## Criteria for article selection
The following criteria were determined to choose articles:
## Results of systematic review
This section is divided into two parts: the first deals with medical imaging modaliti and the second deals with the analysis of medical images. Each revolves around the ma objectives of the systematic review.
## Medical imaging modalities
Medical images play a critical role in assisting health care workers in reaching p tients for diagnosis and treatment. Medical image processing is a set of procedures f extracting clinically useful data from various imaging modalities for diagnosis [bib_ref] Medical image segmentation: A review, Patil [/bib_ref]. N merous medical imaging modalities include ionizing radiation, magnetic resonance, n clear medicine, optical methods, and ultrasound as the media. Each modular media h unique characteristics and responses to the human body's structure [bib_ref] A Survey of Medical Image Classification Techniques, Miranda [/bib_ref]. These modalit serve various purposes, such as obtaining images inside the human body or image sa ples of parts that cannot be seen with the naked eye [bib_ref] Medical Image Analysis using Deep Learning: A Review, Nisa [/bib_ref]. The classification of medi imaging modalities and the main types of imaging methods addressed in the survey studies are illustrated in [fig_ref] Figure 2: Classification of medical imaging modalities [/fig_ref]. The distribution of the forty chosen studies that used different modalities is illu trated in [fig_ref] Table 1: Distribution of studies for different medical imaging modalities [/fig_ref] aging modality, type of disease, and medical databases. The distribution of the forty chosen studies that used different modalities is illustrated in [fig_ref] Table 1: Distribution of studies for different medical imaging modalities [/fig_ref] ; this table shows the detailed distribution of publication references, imaging modality, type of disease, and medical databases.
# Medical image analysis
This section is divided into subsections. These subsections introduce the major medical image analysis methods used in the studies reviewed, including image preprocessing, image segmentation, feature extraction to classification, and evaluation metrics.
## Medical image preprocessing
Image processing is a method for enhancing the quality of an image after eliminating irrelevant image data. Medical images include many irrelevant and unwanted segments. To remove these segments in an image, some preprocessing methods are required. Image preprocessing aims to improve the quality of the images contained in the dataset, which improves the results of segmentation and feature extraction methods [bib_ref] Preprocessing by contrast enhancement techniques for medical images, Perumal [/bib_ref]. This section presents the preprocessing methods of the studies surveyed. One of the most important techniques to improve medical images and remove noise is filtering. One of the most important filters used is the median filter [bib_ref] Classification of Liver Tumor using Modified SFTAbased Multi Class Support Vector Machine, Edwin [/bib_ref] [bib_ref] An Intelligent System for Early Assessment and Classification of Brain Tumor, Keerthana [/bib_ref] [bib_ref] An Efficient Segmentation and Classification System in Medical Images Using Intuitionist Possibilistic..., Chowdhary [/bib_ref] [bib_ref] Advanced Analysis of Cardiac Image Processing Using Hybrid Approach, Kumar [/bib_ref] [bib_ref] A hybrid svm naïve-bayes classifier for bright lesions recognition in eye fundus..., Gharaibeh [/bib_ref] [bib_ref] An Efficient Classification of MRI Brain Images, Assam [/bib_ref] [bib_ref] Prediction of Thyroid Disease(Hypothyroid) in Early Stage Using Feature Selection and Classification..., Riajuliislam [/bib_ref] , with its major benefit being to preserve the edges and remove noise. In [bib_ref] Automated detection of dermatological disorders through image-processing and machine learning, Hasija [/bib_ref] , the authors used the synthetic minority over-sampling technique to generate synthetic samples from minor classes rather than simply replicating them, and they used the changing perspective of images technique to expand the dataset. To generate new images, computer vision techniques such as gray scaling, blurring, enhancing contrast, changing the color channel, sharpening, minimizing noise, and smoothing were used. In [bib_ref] An Intelligent System for Early Assessment and Classification of Brain Tumor, Keerthana [/bib_ref] , the authors described in their study how to utilize pixel-wise interpolation and the modified quadratic transform-based Radon transform algorithm to improve the contrast of the images. In [bib_ref] An Intelligent System for Early Assessment and Classification of Brain Tumor, Keerthana [/bib_ref] [bib_ref] A New Alzheimer's Disease Classification Technique from Brain MRI images, Awasthi [/bib_ref] [bib_ref] Diagnosis of Ophthalmic Diseases in Fundus Image Using various Machine Learning Techniques, Patankar [/bib_ref] , the authors applied the contrast limited adaptive histogram equalization (CLAHE) approach and morphological operations to remove noises and improve the images. In [bib_ref] A Hybrid Diagnosis System for Malignant Melanoma Detection in Dermoscopic Images, Pallavi [/bib_ref] [bib_ref] Lung Cancer Detection using Probabilistic Neural Network with modified Crow-Search Algorithm. Asian..., Sannasi Chakravarthy [/bib_ref] , the authors used a fast adaptive median filter to eliminate the noise from medical images and maintain accurate information. In [bib_ref] A Hybrid Classification and Prediction Methodology for the Diagonosis of Osteoporosis, Kumar [/bib_ref] , binarization and thinning techniques were applied to the selected images to produce better images. In [bib_ref] Deep convolutional neural network based medical image classification for disease diagnosis, Yadav [/bib_ref] [bib_ref] Improved fuzzy clustering with swarm intelligence for medical image analysis, Gholami [/bib_ref] [bib_ref] Study of Fusion of medical images and classification comparison using different kernels..., Kaur [/bib_ref] , the authors improved the noise removal process for medical images to obtain clear images using a Weiner filter for contrast adjustment. In [bib_ref] A Novelty Approach to Retina Diagnosing Using Biometric Techniques With SVM and..., Szymkowski [/bib_ref] , the authors developed two methods to convert the input color images to grayscale, denoise, enhance contrast, and normalize the histogram. The first method utilized only a green channel, while the second method depended on calculating the grayscale by taking an average of red, green, and blue channels computed using certain weights.
In [bib_ref] Alzheimer Disease Detection Based on Deep Neural Network with Rectified Adam Optimization..., Suresha [/bib_ref] [bib_ref] Brain Tumor Detection in MRI Images Using Image Processing Techniques, Hashan [/bib_ref] [bib_ref] A Modified Convolutional Neural Networks For MRI-based Images For Detection and Stage..., Alshammari [/bib_ref] , the authors addressed the normalization method to enhance the visual quality of images. In [bib_ref] An Efficient Segmentation and Classification System in Medical Images Using Intuitionist Possibilistic..., Chowdhary [/bib_ref] the authors used noise removal techniques such as max-min filter, midpoint filter, quantum noise filter, alpha-trimmed mean filter, impulse noise filter, and wavelet thresholding methods for noise removal from images to smoothened them to obtain better images. In [bib_ref] Advanced Analysis of Cardiac Image Processing Using Hybrid Approach, Kumar [/bib_ref] , the authors employed linear and non-linear image processing filters such as mean, Gaussian, Log, and fuzzy filters to eliminate noise from images. In [bib_ref] A hybrid svm naïve-bayes classifier for bright lesions recognition in eye fundus..., Gharaibeh [/bib_ref] , the authors presented the BBHE approach (i.e., Brightness Preserving Bi-Histogram Equalization), a method that decomposes the original image into two sub-images to overcome the drawback of histogram equalization. This is accomplished using a graylevel and histogram equalization approach for each sub-image. The Gradwrap algorithm, addressed in [bib_ref] Alzheimer and Mild Cognitive disease Recognition Using Automated Deep Learning Techniques, Zubair [/bib_ref] , improves the image appearance by removing image distortions. The B1 non-uniformity algorithm corrects the image color and intensity information, and then the N3 bias field correlation is applied after the Gradwrap and B1 non-uniformity algorithms to correct intensity distortion. In [bib_ref] A Machine Learning Approach to Diagnosing Lung and Colon Cancer Using a..., Masud [/bib_ref] , the authors used unsharp masking, a popular image sharpening method, to improve the contrast of histopathological images. In [bib_ref] Breast Cancer Diagnosis by Convolutional Neural Network and Advanced Thermal Exchange Optimization..., Cai [/bib_ref] the Wang-Mendel algorithm was employed to remove noise from the images; this algorithm is one of the most effective approaches for eliminating noise from medical images due to its high speed.
In [bib_ref] SVM based Supervised Machine Learning Framework for Glaucoma Classification using Retinal Fundus..., Parashar [/bib_ref] , the authors analyzed fundus images for image scaling; R, G, and B channel selection; and the preprocessed green channel components were analyzed by the 2D-VMD technique, which has a non-stationary, non-recursive, and completely adaptive decomposi-tion technique for signal image analysis. In [bib_ref] Sa'ad, A. A transfer learning with deep neural network approach for diabetic..., Al-Smadi [/bib_ref] , the authors used image normalization and data over-sampling techniques with data augmentation to enlarge the dataset so that it could be used for deep learning tasks. Data augmentation approaches use a variety of operations to images, including scaling, geometric deformation, noise addition, alterations to the lighting, and image flipping. The authors employed the straightforward and efficient semi-supervised learning technique of pseudo-label to improve the performance of deep neural network models.
## Segmentation techniques
Image segmentation is responsible for recognizing and outlining items of interest in input images. Automatic medical image segmentation aids in the diagnosis of diseases and the identification of pathogens. Image segmentation methods can be divided into machine learning methods such as supervised and unsupervised machine learning and classical segmentation methods such as threshold-based, edge-based, and region-based methods [bib_ref] A State-of-the-Art Survey for Microorganism Image Segmentation Methods and Future Potential, Kulwa [/bib_ref]. This section presents the segmentation methods used by the studies surveyed. In [bib_ref] Classification of Liver Tumor using Modified SFTAbased Multi Class Support Vector Machine, Edwin [/bib_ref] [bib_ref] An Intelligent System for Early Assessment and Classification of Brain Tumor, Keerthana [/bib_ref] [bib_ref] Classification of Lung Diseases Using a Combination of Texture, Shape and Pixel..., Thamke [/bib_ref] [bib_ref] A Hybrid Diagnosis System for Malignant Melanoma Detection in Dermoscopic Images, Pallavi [/bib_ref] [bib_ref] A Hybrid Classification and Prediction Methodology for the Diagonosis of Osteoporosis, Kumar [/bib_ref] [bib_ref] Lung Cancer Detection using Probabilistic Neural Network with modified Crow-Search Algorithm. Asian..., Sannasi Chakravarthy [/bib_ref] , the authors used thresholding. In this method, the image is partitioned into its foreground and background depending on the threshold value. In [bib_ref] Classification of Liver Tumor using Modified SFTAbased Multi Class Support Vector Machine, Edwin [/bib_ref] [bib_ref] Brain Tumor Detection in MRI Images Using Image Processing Techniques, Hashan [/bib_ref] , the authors applied the contour technique; active contour is an active segmentation model that separates the pixels of interest from an image using energy forces and limits. In [bib_ref] CDCT: CT Scan Images based on Mechanism for Lung Cancer Detection, Sharma [/bib_ref] [bib_ref] Advanced Analysis of Cardiac Image Processing Using Hybrid Approach, Kumar [/bib_ref] , the authors suggested using watershed segmentation techniques; this approach is used to separate different objects. The study [bib_ref] Normal And Abnormal Detection For Knee Osteoarthritis Using Machine Learning Techniques, Yousuf Bhat [/bib_ref] [bib_ref] Advanced Analysis of Cardiac Image Processing Using Hybrid Approach, Kumar [/bib_ref] employed Otsu's threshold approach to extract the object; this approach finds an optimum threshold automatically. Because of its simple calculation, Otsu's is the most successful technique for image thresholding. The authors in [bib_ref] A Novelty Approach to Retina Diagnosing Using Biometric Techniques With SVM and..., Szymkowski [/bib_ref] used the Gaussian matched filter approach with binarization realized with local entropy thresholding. This method allowed them to acquire more exact results than using only one binarization threshold. Additionally, the authors employed the crossing number algorithm or minutiae detection to extract minutiae and then automatically count their number.
The study [bib_ref] An Efficient Classification of MRI Brain Images, Assam [/bib_ref] adopted a color space-based method for mammography image segmentation, followed by mathematical morphology. Post-processing mathematical morphology improved performance, including filling, closure, and other operations.
## Feature extraction techniques
Feature extraction is the process of extracting meaningful data from raw data. It is crucial in image processing as it enhances the image's quality by reducing the dimensionality of the image, extracting the unique features, and transforming the input data into a set of features that are used for classification purposes. There are various features such as color, texture, and shape, and each type has several methods to extract features from medical images [bib_ref] A Detailed Survey On Feature Extraction Techniques In Image Processing For Medical..., Kumar [/bib_ref]. The distribution of the chosen studies that used different feature extraction and reduction methods is illustrated in [fig_ref] Table 2: Distribution of studies for different feature extraction methods [/fig_ref] ; this table comprises the detailed distribution of publication references, the type of features, and the method used.
## Classification techniques
Image classification is a difficult task in image analysis. The primary purpose of medical image classification is to accurately establish which parts of the human body are infected with the disease [bib_ref] Medical Image Classification Based on Deep Features Extracted by Deep Model and..., Lai [/bib_ref]. This study reviews the chosen studies that used the newest classification techniques for medical image classification. The study [bib_ref] An MR brain images classification technique via the Gaussian radial basis kernel..., Neffati [/bib_ref] proposed a new classifier depending on KPCA and SVM. The support vector machine with kernel principal component analysis (SVM-KPCA) approach was designed to classify the images, and the proposed method achieved effective results, which gave 100% accuracy, 100% sensitivity, and 100% specificity. In the study [bib_ref] Analysis of Brain MRI Tumor Detection and Classification using Hybrid Approach, Varun Jain [/bib_ref] , the authors suggested a hybrid approach consisting of a radial basis function neural network (RBFNN) to classify brain MRI images, the accuracy of this model varied from 80% to 90%. [bib_ref] Image based skin disease detection using hybrid neural network coupled bag-of-features, Chakraborty [/bib_ref] Several key points such as features from which it creates a descriptor for that point SIFT to detect key (interest) points and feature extraction. Bag-of-features concept to decrease the number of key points. [bib_ref] Diagnosis of Alzheimer's Disease Using Machine Learning, Lodha [/bib_ref] Cognitive and biological features The full volume of the brain is extracted from MRI images and other cognitive and biological features. The study [bib_ref] Image based skin disease detection using hybrid neural network coupled bag-of-features, Chakraborty [/bib_ref] developed a neural-based detection system by employing skin imaging for two different skin disorders. The ANN was trained using the non-dominated sorting genetic algorithm-II, a well-known multi-objective optimization technique (NN-NSGA-II). The proposed model was compared to two well-known metaheuristic-based classifiers, NN-PSO (ANN trained with PSO) and NN-CS (ANN trained with Cuckoo Search). The proposed bag-of-features enabled the N-NSGA-II model and obtained 90.56% accuracy, 88.26% precision, 93.64% recall, and 90.87% F-measure with the experimental data, indicating its superiority over other models. The study [bib_ref] Automated detection of dermatological disorders through image-processing and machine learning, Hasija [/bib_ref] applied multiple artificial intelligence (AI) techniques, such as the convolutional neural network and support vector machine, which were combined with image processing tools to construct a superior structure; the accuracy attained after training with CNN alone was approximately 91%, which was raised to approximately 95.3% when combined with the SVM. In [bib_ref] CDCT: CT Scan Images based on Mechanism for Lung Cancer Detection, Sharma [/bib_ref] , the authors developed a new SVM-FA (support vector machine optimized with firefly technique) classifier for diagnosing lung cancer in CT images where the SVM classifier, optimized with the firefly technique, was applied to the preprocessed data. A comparative analysis was conducted between the proposed work, traditional work, and the SVM classifier to assess the competence level of the proposed SVM-FA technique. The suggested work was successful and efficient, achieving an accuracy of 96 % and specificity of 83.3%.
The study [bib_ref] Alzheimer Disease Detection Based on Deep Neural Network with Rectified Adam Optimization..., Suresha [/bib_ref] applied a deep neural network (DNN) with the rectified Adam optimizer to detect Alzheimer's disease in MRI images. The experimental outcomes showed that DNN with the rectified Adam optimizer outperformed the existing work, landmarkbased features with the SVM classifier, with a 16% classification accuracy. In a study [bib_ref] A hybrid svm naïve-bayes classifier for bright lesions recognition in eye fundus..., Gharaibeh [/bib_ref] , the authors addressed a combination of MLC (maximum likelihood classifier) and SVM (support vector machine) classifiers for the classification and diagnosis of DR (diabetic retinopathy) disease in the fundus images. The researchers utilized these classifiers to raise the performance level, and the suggested approach demonstrated high accuracy (98.60%), sensitivity (99%), and specificity (99%). In [bib_ref] An Efficient Classification of MRI Brain Images, Assam [/bib_ref] , the authors introduced an individual classifier called: feed-forward ANN (FF-ANN) and two hybrid classifiers, namely: random subspace with random forest (RSwithRF) and random subspace with Bayesian network (RSwithBN), for the classification of MRI brain images. The proposed system showed that ANN and hybrid classification approaches are the most appropriate for classification because of their high accuracy rates and achieved an accurate classification of 95.83%, 97.14%, and 95.71%, respectively.
In [bib_ref] Diagnosis of Ophthalmic Diseases in Fundus Image Using various Machine Learning Techniques, Patankar [/bib_ref] , the authors used various techniques, such as machine learning and different deep learning models, to predict various ophthalmic diseases; the study showed that the efficiency of each strategy varied depending on the input dataset and based on the many symptoms of each disease.
## Metric evaluation
Accuracy, specificity, and sensitivity are the most prevalent metrics utilized in most disease diagnosis applications for humans. The number of true positive (TP), true negative (TN), false positive (FP), and false-negative (FN) samples are used to calculate these measures [bib_ref] New machine learning method for image-based diagnosis of COVID-19, Elaziz [/bib_ref]. The sensitivity determines how many positive samples were identified (TP), whereas the specificity determines how many negative samples were identified (TN). The ratio of correctly recognized samples to the total number of samples is used to calculate classification accuracy [bib_ref] New machine learning method for image-based diagnosis of COVID-19, Elaziz [/bib_ref]. Other metrics such as precision, recall, F1 score, and AUC must be used in conjunction with these metrics [bib_ref] Intelligent Pneumonia Identification From Chest X-Rays: A Systematic Literature Review, Khan [/bib_ref]. [fig_ref] Table 3: Frequency count of performance metrics utilized in every chosen study [/fig_ref] illustrates the number of times each performance metric was utilized in each study.
## Machine learning techniques
This study classified machine learning (ML) segmentation and classification techniques as supervised and unsupervised machine learning. Supervised learning algorithms generate mathematical models using a set of labeled data (images) which are utilized for training, examples of supervised machine learning algorithms are K-nearest neighbors (K-NN), support vector machines (SVM), decision trees (DT), linear regression, logistic regression, random forest (RF), artificial neural networks (ANN), gradient boosting, and naïve Bayes models. Unsupervised learning algorithms create mathematical models based on a set of data that only contains inputs and no required output labels. The algorithms identify patterns in the data and classify them. Examples of unsupervised machine learning algorithms are K-mean clustering, hierarchical clustering, Apriori algorithm, principal component analysis (PCA), and fuzzy C-means (FCM).
In the surveyed studies, several techniques of machine learning (ML) in segmentation and classification were deployed. The study [bib_ref] A New Alzheimer's Disease Classification Technique from Brain MRI images, Awasthi [/bib_ref] compared the IPFCM (intuitionist possibilistic fuzzy C-mean) method with the hybridization of the negative function of the intuitionistic and the negative function of the possibilistic for mammogram image segmentation with conventional segmentation methods such as the Otsu algorithm, FCM (fuzzy C-mean) clustering, IFCM (intuitionist fuzzy C-mean) clustering, and PFCM (possibilistic fuzzy C-mean). Moreover, they found that this proposed method was the best approach. In studies [bib_ref] Improved fuzzy clustering with swarm intelligence for medical image analysis, Gholami [/bib_ref] [bib_ref] Advanced Analysis of Cardiac Image Processing Using Hybrid Approach, Kumar [/bib_ref] [bib_ref] MRI Image Based Classification Model for Lung Tumor Detection Using Convolutional Neural..., Kumar [/bib_ref] , the authors used the fuzzy clustering method. Fuzzy clustering is the most extensively used for image segmentation because of its benefits compared with traditional clustering methods, which includes: making regions more homogeneous, decreasing the number of erroneous spots, reducing noise sensitivity, and removing noisy regions [bib_ref] Improved fuzzy clustering with swarm intelligence for medical image analysis, Gholami [/bib_ref]. In [bib_ref] A hybrid svm naïve-bayes classifier for bright lesions recognition in eye fundus..., Gharaibeh [/bib_ref] , the authors introduced the MSW-FCM technique (modified spatial weighted fuzzy C-means) for accurately segmenting blood vessels in retinal fundus images.
The distribution of the chosen studies that used the most popular machine learning techniques to diagnose human body disease is illustrated in [fig_ref] Table 4: Distribution of studies for the most common ML classification methods [/fig_ref]. This table comprises the detailed distribution of publication references, techniques used, classified tasks, and classification accuracy results. [bib_ref] SVM based Supervised Machine Learning Framework for Glaucoma Classification using Retinal Fundus..., Parashar [/bib_ref] Multi-stage classifier (SVM) Glaucoma classification 91%
## Deep learning
Deep learning (DL) is well-known for its performance in image segmentation and classification models [bib_ref] Deep and machine learning techniques for medical imaging-based breast cancer: A comprehensive..., Houssein [/bib_ref]. Convolution neural networks (CNN) are the most extensively utilized deep learning approach in the articles reviewed. CNN is an important image processing approach that allows for accurately classifying aberrant and normal samples [bib_ref] Automated breast ultrasound lesions detection using convolutional neural networks, Yap [/bib_ref]. CNN employs a layered perceptron-driven architecture composed of fully connected networks in which every neuron in one layer is coupled to all neurons in the subsequent layers. The input images, an in-depth feature extractor, and a classifier are the three main components of a CNN [bib_ref] Intelligent Pneumonia Identification From Chest X-Rays: A Systematic Literature Review, Khan [/bib_ref]. There are three kinds of layers in CNN, each of which performs a different function: (1) convolutional, (2) pooling, and (3) fully connected. The convolutional layer extracts the characteristics of the structure. The fully connected layer then decides which class the current input belongs to, depending on the retrieved features. Then, the pooling layer is responsible for shrinking feature maps and network parameters. Transfer learning algorithms can increase CNN performance in the case of limited input data [bib_ref] ImageNet classification with deep convolutional neural networks, Krizhevsky [/bib_ref]. A CNN can be created from scratch using an existing pre-trained network without retraining or fine-tuning a pre-trained network on a target dataset [bib_ref] Intelligent Pneumonia Identification From Chest X-Rays: A Systematic Literature Review, Khan [/bib_ref].
According to this review, some research publications used distinct deep neural network architectures; [fig_ref] Table 5: Summary of articles that included deep learning in detecting diseases [/fig_ref] summarizes studies that used deep learning in disease diagnosis.
The study [bib_ref] Combining convolutional and recurrent neural networks for Alzheimer's disease diagnosis using PET..., Cheng [/bib_ref] suggested a new classification structure that depends on a combination of 2D CNN and recurrent neural networks (RNN) that learn the properties of 3D PET images by decomposing them into a series of 2D slices. The intra-slice characteristics are captured using hierarchical 2D CNNs, while the inter-slice features are extracted using the gated recurrent unit (GRU) of an RNN for final classification. The experimental outcomes showed that the suggested method has promising performance for Alzheimer's disease (AD) diagnosis. The authors in [bib_ref] Deep convolutional neural network based medical image classification for disease diagnosis, Yadav [/bib_ref] applied CNN-based algorithm on a chest X-ray dataset to detect pneumonia. Three approaches were examined, a linear support vector machine classifier with local rotation and orientation-free features, transfer learning on two CNN models: Visual Geometry Group, i.e., VGG16 and InceptionV3, and a capsule network training from scratch. Data augmentation is a data preprocessing technique applied to all three approaches, the outcomes revealed that data augmentation is an effective technique for all three algorithms to improve performance and efficiency. In [bib_ref] Alzheimer Disease Detection Based on Deep Neural Network with Rectified Adam Optimization..., Suresha [/bib_ref] , the authors proposed a successful approach for predicting the probability of brain cancers in MRI images using CNNs and the (Adam) optimizer algorithm. An adaptive moment estimation (Adam) optimizer has been introduced to expedite training the network and evaluate the model to attain maximum accuracy. The study [bib_ref] Alzheimer and Mild Cognitive disease Recognition Using Automated Deep Learning Techniques, Zubair [/bib_ref] introduced a new deep learning-based CNN model created by the Bayesian optimization algorithm for classifying Alzheimer's disease, mild cognitive impairment (MCI), and cognitively normal (CN) in MRI images. The proposed method gives extraordinary outcomes compared to the existing techniques. The study [bib_ref] Breast Cancer Diagnosis by Convolutional Neural Network and Advanced Thermal Exchange Optimization..., Cai [/bib_ref] compared a new optimized version of CNN and a new, improved metaheuristic, named the advanced thermal exchange optimizer for the detection of breast cancers with three different techniques including multilayer perceptron (MLP), multiple instances (MI), and transfer learning (TL), which were applied in the MIAS mammography database to demonstrate the superiority of proposed method. In a study [bib_ref] A Modified Convolutional Neural Networks For MRI-based Images For Detection and Stage..., Alshammari [/bib_ref] , the authors suggested using the ML approach in the training process to create a prediction model by training the CNN algorithm in addition to using the "Adam" optimizer from the Python Keras optimization library that has an initial learning rate of 0.001. The evaluation was carried out after partitioning the data into 80% and 20% for training and evaluation, respectively, to compute the accuracy of classification and loss of model over a set number of 10 epochs, the proposed algorithm gave good results. In the study [bib_ref] Sa'ad, A. A transfer learning with deep neural network approach for diabetic..., Al-Smadi [/bib_ref] , the authors developed a transfer learning-based technique to determine the severity of diabetic retinopathy; the proposed model was a deep learning model that combined multiple pre-trained image classification CNN models with the global average pooling (GAP) technique. The accuracy of the model attained 82.4% quadratic weighted transfer learning kappa (QWK). In [bib_ref] Integration of CNN, CBMIR, and visualization techniques for diagnosis and quantification of..., Mohagheghi [/bib_ref] , the authors proposed a diagnosis system based on deep neural networks and image retrieval method. Transfer learning and hashing functions increased the CNN performance and image retrieval algorithms. The proposed system attained an accuracy of 97% for CNN and a content-based medical image retrieval (CBMIR) method. The study [bib_ref] Modality classification and concept detection in medical images using deep transfer learning, Singh [/bib_ref] compared the effectiveness of modern CNN models for the task of modality classification and reported the superiority of a deep learning-based method over classic feature engineering approaches based on multi-label learning algorithms. The experimental results demonstrated that deep learning is more efficient than traditional methods and produced better and more robust feature representations when compared to handcrafted feature extraction approaches. The findings showed that deep transfer learning techniques work well in the medical field, where data is scarce. The Google Inception-v3 model performed the best when it came to classifying medical picture modalities. Except for VGG-16 and Res-Net-50, the other models behaved similarly to Inception-v3.
Generally, the studies utilized accuracy, specificity, sensitivity, precision, recall, f1 score, and AUC as evaluation metrics. According to the conclusions of the studies, deep learning algorithms achieved good results in most of the evaluation metrics (as in [fig_ref] Table 5: Summary of articles that included deep learning in detecting diseases [/fig_ref].
## Diseases diagnosis system
Image processing has been extensively employed in various illness diagnosis procedures (human, animal, and plant), helping professionals select the appropriate treatment. In diagnosing human diseases, image processing techniques play an essential role. They can be used to identify disease signs (on the skin, for example) or in molecular research using microscope images that show the anatomy of the tissues. The disease diagnosis systems comprise stages and methods for diagnosis. It starts from the first stage, which is the stage of collecting images from different sources, either through a database available online or collecting the images from different sources, i.e., images of patients available on the internet or from a specific hospital. Then the preprocessing image stage as different filtering methods are applied to improve the image, and then the segmentation stage follows by isolating the regions of interest and extracting the important features in the feature extraction stage. At the end of the process, the input image is classified and metrics for evaluating the effectiveness of disease diagnosis techniques are applied. The general system stages for the diagnosis of any disease in image processing are illustrated in [fig_ref] Figure 3: Conceptual map of disease diagnosis system [/fig_ref]. This figure shows a conceptual map that explains concepts linked to disease diagnosis steps and a descriptive brief of the major evaluation metrics.
In this study, several disease diagnosis methods were studied that used different image treatment and classification strategies, which dealt with diagnosing human diseases, to examine the new and important methods that addressed in the studies. In this study, several disease diagnosis methods were studied that used different age treatment and classification strategies, which dealt with diagnosing human diseas to examine the new and important methods that addressed in the studies.
## Discussion and future directions
The growing interest in employing medical image processing approaches and techniques, such as ML and DL, may reduce the doctor's workload and repetitive a monotonous procedures to diagnose and analyze patient data and images [bib_ref] Learning About Machine Learning to Create a Self-Driving Echocardiographic Laboratory: Technical Considerations, D'hooge [/bib_ref]. In t section, we discuss the state-of-the-art approaches and datasets used to diagnose hum diseases through an answer to research questions and future directions.
## Answer to research questions
The data were extracted from the studied articles as they were explained and clarif to answer the research questions.
"Q1: What are the modalities of medical imaging?" The methods of medical imag were identified in the articles, where many modern imaging methods for diseases ha been used, which help in diagnosing diseases in the early stages. The MRI imaging m dality was used in fourteen articles in the diagnosis of brain diseases, lung diseases, c diovascular diseases, and cardiac attacks; the CT imaging modality was used in eight ticles in the diagnosis of lung diseases, liver tumors, and bone diseases. The X-Ray m dality was utilized in seven articles to diagnose osteoarthritis disease, pneumonia disea and breast cancer. Five articles used retina fundus images to diagnose eye illnesses retina disease diagnosis. PET imaging was utilized in four articles to diagnose Alzheime disease, and dermoscopy skin imaging was used to diagnose skin problems.
"Q2: What is the task of medical image processing and analysis?" In the analysis the articles, it was identified that the task of the analysis of medical images and process it to diagnose several diseases in different parts of the human body to assist the patien detecting the disease and treating it in the early stage.
## Discussion and future directions
The growing interest in employing medical image processing approaches and AI techniques, such as ML and DL, may reduce the doctor's workload and repetitive and monotonous procedures to diagnose and analyze patient data and images [bib_ref] Learning About Machine Learning to Create a Self-Driving Echocardiographic Laboratory: Technical Considerations, D'hooge [/bib_ref]. In this section, we discuss the state-of-the-art approaches and datasets used to diagnose human diseases through an answer to research questions and future directions.
## Answer to research questions
The data were extracted from the studied articles as they were explained and clarified to answer the research questions.
"Q1: What are the modalities of medical imaging?" The methods of medical imaging were identified in the articles, where many modern imaging methods for diseases have been used, which help in diagnosing diseases in the early stages. The MRI imaging modality was used in fourteen articles in the diagnosis of brain diseases, lung diseases, cardiovascular diseases, and cardiac attacks; the CT imaging modality was used in eight articles in the diagnosis of lung diseases, liver tumors, and bone diseases. The X-Ray modality was utilized in seven articles to diagnose osteoarthritis disease, pneumonia disease, and breast cancer. Five articles used retina fundus images to diagnose eye illnesses for retina disease diagnosis. PET imaging was utilized in four articles to diagnose Alzheimer's disease, and dermoscopy skin imaging was used to diagnose skin problems.
"Q2: What is the task of medical image processing and analysis?" In the analysis of the articles, it was identified that the task of the analysis of medical images and processing it to diagnose several diseases in different parts of the human body to assist the patient in detecting the disease and treating it in the early stage.
"Q3: Which medical image processing methods are most used in diagnostic systems?" Through the analysis of the articles, we discovered that by using filtering techniques, we may improve the initial image and get a more exact detection of the ROI borders. Noise can be reduced by smoothing with a low pass or median filtering, while edge sharpening and increased contrast can help to keep the ROIs' borders. Color normalization, such as histogram equalization or specification, can boost contrast. A variety of transforms might be used to extract important features. Fourier and wavelet transform can be utilized to locate conditions in a domain other than the spatial one. Because the borders of the object or regions containing the crucial information for the diagnosis must be identified correctly, image segmentation is one of the most essential phases in a disease diagnosis procedure. The simplest but effective method is gray-level segmentation utilizing a single, multiple, or more advanced (Otsu) criteria. Several human illness diagnosis approaches have used active contour detection and modifications, such as a snake. Several clustering approaches were applied in the segmentation process, such as K-mean and fuzzy C-means, to separate the ROIs. The articles used many methods to extract the important features from objects. We found that the texture features were the most used in analyzing the articles. It was obvious from the analysis of the articles that the most popular classification method used in articles was a neural network with all types and SVM, where it was used alone or with another technique as a hybrid system for the classification of diseases. The rest of the classification techniques examined in this review such as K-means, K-NN, decision trees, naïve Bayes, random forest, logistic regression, and gradient boosting. All the classification or clustering methods can be exploited in the last stage of the diagnosis application to create a new diagnosis system to identify disease.
"Q4: What diagnostic techniques have been adopted and developed?" Analyzing the articles, we found that some of the articles adopted new hybrid methods for classification and diagnosis of diseases, such as using the SVM method with KPSA, SVM with FA, SVM with MLC, as well as using it with fuzzy. Moreover, the combination of random forest (RF) with random subspace (RS) and the networks CNN with RNN and with optimizers were used to create systems for disease diagnosis.
"Q5: Is the system that has been adopted or designed capable of producing good results?" The main classification methods discussed in the articles achieved efficient and good results in diagnosing diseases. The accuracy of SVM ranged between 88% and 100% in the several human disease diagnosis applications examined, naïve Bayes achieved accuracy between 78% and 94%, decision trees ranged between 82% and 99%, and random forest ranged between 80% and 99%. The different kinds of neural networks achieved accuracy between 73% and 97%, and the accuracy of CNN ranged between 86% and 99.3%. Finally, the accuracy of the K-nearest neighbor ranged between 73% and 95.5% in the referenced approaches.
## Future directions
It takes great effort to improve the performance of classification or diagnosis of diseases using multiple methods of medical images. In future work, we will try to present new research directions that can be further exploited in disease diagnosis through medical image processing techniques. We will compare the most common methods that are used in the diagnosis systems and select the most effective methods that introduce higher accuracy to the diagnosis for the medical image database that we use to build a new diagnostic system for diseases. The future research directions are discussed briefly as follows:
A. Study and analysis of the best and most common classification and diagnosis methods. B. Experimenting with a set of medical data to compute the classification accuracy of the methods used.
C. Comparison of classification accuracy of these methods to identify the methods that are highly accurate.
D. Building a new diagnostic method by deriving a hybrid method, modifying a previous method, or combining previous methods.
The current study improved our knowledge of finding the best techniques that can be beneficial to our future research. We hope that this systematic literature review will also be useful to other researchers in their endeavor to improve disease diagnosis through medical image processing techniques. Due to our comparisons, the most common methods used in diagnosis systems have been discussed, making it easier to select effective methods that introduce higher accuracy to diagnosis methods using medical databases.
# Conclusions
Medical imaging plays a critical role in inspecting and diagnosing human diseases. For diagnosis, many algorithms based on diverse methodologies have been created. As a result, disease detection has become an important topic in medical image processing and medical imaging research.
This review studied the articles on disease diagnosis published between 2017 and 2021. Overall, forty articles were analyzed from specialized academic repositories. The review focused on six factors: datasets used, various medical imaging modalities, image preprocessing techniques, image segmentation techniques, feature extraction techniques, classification approaches, and performance metrics, which were used to build and evaluate the disease diagnostic models. In addition, this systematic review highlighted a variety of AI techniques and presented a comprehensive study by exploring new diagnosis techniques, disease diagnosis issues, provided a variety of insightful information (such as the use of ML and DL), and an evaluation for each study. We aimed to address our study questions about effective diagnosis methodologies and discover the solutions proposed by many researchers in the diagnosis of diseases based on this. It was discovered that developing a new disease diagnosis method is quite important.
According to the findings of this SLR, researchers have adopted various methods to classify medical images associated with multiple disease diagnoses. These methods have shown promising results in terms of accuracy, cost, and detection speed. We found after analyzing the forty articles that the best preprocessing technique was the median filter, which was used in many studies, as has proven its ability to reduce noise and preserve the boundaries of the object. Regarding segmentation approaches, threshold techniques were the most used to extract the lesion from an image. Threshold-based approaches are the most often utilized among all the traditional methods, according to classic review publications, due to their applicability for numerous segmentation issues in medical images [bib_ref] A Systematic Literature Review of Medical Image Analysis Using Deep Learning, Buettner [/bib_ref]. The methods for extracting features from medical images depend on the images used by analyzing the articles, we found that extracting texture features gave the best results. As for the applied classification methods, we found that the support vector machine method gave the best result in classification, as its accuracy in [bib_ref] An MR brain images classification technique via the Gaussian radial basis kernel..., Neffati [/bib_ref] reached 100% when it was used with KPC (support vector machine with kernel principal component analysis (SVM-KPCA)) approach. Thus, this method achieved the best accuracy and the best performance in diagnosis. Among the machine learning-based approaches whose associated works and analyses are presented, the supervised learning methods, notably the neural network-based methods, were the most widely employed, with different kinds of neural networks used to identify various diseases.
We also discovered that deep learning utilizing the CNN network has unique skills and advances in recognizing and classifying medical images, particularly those connected to breast, lung, and brain cancer. Other common classification approaches comprise fuzzy clustering, K-NN, K-means, decision trees, random forests, and other prominent classification algorithms. The most utilized measures were accuracy, sensitivity, and specificity.
This study aimed to propose future research directions by focusing on imaging modalities, techniques, and procedures used in the reviewed articles. Furthermore, this publication will aid in the development of new research that assesses and compares various medical image processing and analysis techniques. The findings provided in this SRL reveal that tremendous progress has been made in medical image processing over the last five years. In addition, the goal of this SRL was to undertake a detailed analysis of research achievements linked to the usage of medical image processing techniques applied to medical databases to know the current state-of-the-art techniques.
Author Contributions: The concept of the article was proposed by N.P., the data resources and validation were contributed by B.M.R., the formal analysis, investigation, and draft preparation were performed by B.M.R. The supervision and review of the study were headed by N.P. The final writing
[fig] Figure 1: Flowchart of the process for article selection. [/fig]
[fig] Figure 2: Classification of medical imaging modalities. [/fig]
[fig] Figure 3: Conceptual map of disease diagnosis system. [/fig]
[table] Table 1: Distribution of studies for different medical imaging modalities. Open Access Series of Imaging Studies (OASIS) website. The dataset contained normal brains and seven types of pathological brains with a total of 226 images (38 normal brains and 188 pathological brains). CT scan image dataset collected from patients (age ranges from 35 to 75). The datasets contained 400 images (100 normal, 100 pleural effusion, 100 bronchitis, and 100 emphysema). [/table]
[table] Table 2: Distribution of studies for different feature extraction methods. DWT transforms to extract features. The kernel PCA (KPCA) technique for feature reduction. Varun Jain et al. (2017)[68] Texture features DWT transform for feature extraction. PCA technique for diminishing the number of features. [/table]
[table] Table 3: Frequency count of performance metrics utilized in every chosen study. [/table]
[table] Table 4: Distribution of studies for the most common ML classification methods. [/table]
[table] Table 5: Summary of articles that included deep learning in detecting diseases. [/table]
[bib_ref] Multiclass Classification of Brain Cancer with Machine Learning Algorithms, Erkal [/bib_ref] [bib_ref] Brain Tumor Detection in MRI Images Using Image Processing Techniques, Hashan [/bib_ref] [bib_ref] Combining convolutional and recurrent neural networks for Alzheimer's disease diagnosis using PET..., Cheng [/bib_ref] [bib_ref] Deep convolutional neural network based medical image classification for disease diagnosis, Yadav [/bib_ref] [bib_ref] Automatic Classification of Gastrointestinal Diseases Based on Machine Learning Techniques, Rabi [/bib_ref] [bib_ref] An Efficient Segmentation and Classification System in Medical Images Using Intuitionist Possibilistic..., Chowdhary [/bib_ref] [bib_ref] Optimized CNN-based diagnosis system to detect the pneumonia from chest radiographs, Aledhari [/bib_ref] [bib_ref] Prediction of Thyroid Disease(Hypothyroid) in Early Stage Using Feature Selection and Classification..., Riajuliislam [/bib_ref] |
Investigation on Dynamic Calibration for an Optical-Fiber Solids Concentration Probe in Gas-Solid Two-Phase Flows
This paper presents a review and analysis of the research that has been carried out on dynamic calibration for optical-fiber solids concentration probes. An introduction to the optical-fiber solids concentration probe was given. Different calibration methods of optical-fiber solids concentration probes reported in the literature were reviewed. In addition, a reflection-type optical-fiber solids concentration probe was uniquely calibrated at nearly full range of the solids concentration from 0 to packed bed concentration. The effects of particle properties (particle size, sphericity and color) on the calibration results were comprehensively investigated. The results show that the output voltage has a tendency to increase with the decreasing particle size, and the effect of particle color on calibration result is more predominant than that of sphericity.
# Introduction
The measurements of solids concentration are essential to understand the gas-solid flow behavior in fluidized beds, blow tanks, pneumatic conveying lines, and other multiphase flow systems. A detailed knowledge of solids concentration profile is critical to the accurate design and valid modeling of these OPEN ACCESS systems. Many techniques have been carried out for measuring solids concentration: X-ray or γ-ray absorption, laser Doppler anemometry, acoustics methods, capacitance probes, optical-fiber probes, and so on [bib_ref] Measurements of solids concentration and axial solids velocity in gas-solid two-phase flows, Nieuwland [/bib_ref] [bib_ref] Capacitance probes for solids volume concentration and velocity measurements in industrial fluidized..., Wiesendorf [/bib_ref] [bib_ref] Measurement techniques in fluidized beds, Werther [/bib_ref] [bib_ref] Measuring the gas-solids distribution in fluidized beds-A review, Van Ommen [/bib_ref] [bib_ref] Integration of ECT measurements with hydrodynamic modelling of conventional gas-Solid bubbling bed, Makkawi [/bib_ref] [bib_ref] A comparative study between electrical capacitance tomography and time-resolved X-raytomography, Rautenbach [/bib_ref] [bib_ref] Electrical capacitance volume tomography: Design and applications, Wang [/bib_ref]. Optical-fiber probes have been widely used in recent years for the determination of velocity and concentration of particles in gas-solid flow systems [bib_ref] Measurements of solids concentration and axial solids velocity in gas-solid two-phase flows, Nieuwland [/bib_ref] [bib_ref] Capacitance probes for solids volume concentration and velocity measurements in industrial fluidized..., Wiesendorf [/bib_ref] [bib_ref] Measurement techniques in fluidized beds, Werther [/bib_ref] [bib_ref] Effects of particle properties on flow structure in a 2-D circulating fluidized..., Xu [/bib_ref] [bib_ref] Local solid concentration measurement by fibre optics: Application to circulating fluidized beds, Saberi [/bib_ref] [bib_ref] Dual optical fibre measurements of the particle concentration in gas/solid flows, Rundqvist [/bib_ref] [bib_ref] Solids concentration profiles and pressure gradient distributions, Zhang [/bib_ref] [bib_ref] Characterization of fluidization behavior in the bottom region of CFB risers, Zhu [/bib_ref] [bib_ref] Calculation of optic fibres calibration curves for the measurement of solids volume..., Amos [/bib_ref] [bib_ref] Comparison of fibre optical measurements and discrete element simulations for the study..., Link [/bib_ref] [bib_ref] Optoelectronic technique for the characterization of high concentration gas-solid suspension, Cutolo [/bib_ref] [bib_ref] Fiber-optical sensors: Basics and applications in multiphase reactors, Li [/bib_ref] [bib_ref] Measurements on the Local Solids Concentration in the Lower Part of a..., Ximei [/bib_ref] , which have the advantages of high sensitivity, fast response, high signal-to-noise ratio, large dynamic range, small volume and light weight, fire and shock resistance and corrosion proof, freedom from disturbance by temperature, humidity, electrostatics and electromagnetic fields, and suitability for remote transmission and multi-channel detection. The application of optical-fiber probes to the solids concentration measurement is based on the principle that the particles in the fluid produce scattering of incident light. There are two different arrangements of optical-fiber probes [bib_ref] The use of optic fiber probes for the measurement of dilute particle..., Matsuno [/bib_ref] : transmission-type probe and reflection-type probe, as shown in [fig_ref] Figure 1: Two different types of optic fiber probes[19] [/fig_ref]. As for the transmission-type probe, which is based on the forward scattering of particles against the incident light, the object to be measured is located between the two probe tips and the light input and output are coaxial. The effective measurement volume is dependent on the distance between the two probe tips L, diameter and numerical aperture of the probe. The output signal is independent of the chromaticness of particles, which means that a single white or black particle may produce the same output signals. As for the reflection-type probe, which is based on the back scattering of incident light by particles, it has only one tip, where the projecting and receiving fibers are intermingled or in rows, in parallel or crossed. The effective measurement volume is dependent on the diameter, numerical aperture, overlap region of the capture angles and the optic sensitivity of the photoelectric converter. The output signals depend on the chromaticness and reflectivity of the particles. The transmission-type probe is restricted to relatively low solids concentration and considered to cause strong disturbance of the flow structure compared to the reflection-type probe [bib_ref] Calculation of optic fibres calibration curves for the measurement of solids volume..., Amos [/bib_ref] [bib_ref] Optical fiber measurements of particle concentration in dense suspensions: Calibration and simulation, Lischer [/bib_ref]. The reflection-type probe may be used over the entire range of particle concentration, from extremely dilute flows to the fixed bed state [bib_ref] Application of fiber optic reflection probes to the measurement of local particle..., Herbert [/bib_ref]. Thus, more attention has been paid to the reflection-type optical-fiber probe.
According to the ratio of particle diameter to fiber diameter, Matsuno et al. [bib_ref] The use of optic fiber probes for the measurement of dilute particle..., Matsuno [/bib_ref] classified the reflection-type probes into two categories. shows the optical-fiber probe with fiber diameter larger than particle diameter, the output signals are all generated by the reflected light from the particles existing within the measuring area. The integrated values of the output signal can be correlated with the particle concentration by any calibration method, and the instantaneous concentration can be obtained. shows the optical-fiber probe with fiber diameter smaller than the particle diameter. The output signals from the light receiver are converted into pulses at some threshold level V s , and the pulse count corresponds to the number of particles. The particle velocity must be known in order to convert pulses to concentration. Only the average concentration can be measured if the flow field fluctuates. . Two categories of reflection-type probe [bib_ref] The use of optic fiber probes for the measurement of dilute particle..., Matsuno [/bib_ref] [bib_ref] Novel multifunctional optical-fiber probe: I. Development and validation, Liu [/bib_ref].
[formula] (a) (b) [/formula]
For the measurement of solids concentration, the emitted light reflected by the moving particles is magnified by a photo-multiplier and converted into voltage signals. The optical-fiber solids concentration probe should be calibrated to obtain a relationship between the output voltages and the solids concentration, which depends on the particle size distribution, concentration and optical properties of the solids and the surrounding fluid. The reliability of the measurements is strongly affected by the accuracy of the calibration method [bib_ref] Local solid concentration measurement by fibre optics: Application to circulating fluidized beds, Saberi [/bib_ref] [bib_ref] Fiber-optical sensors: Basics and applications in multiphase reactors, Li [/bib_ref] [bib_ref] Application of fiber optic reflection probes to the measurement of local particle..., Herbert [/bib_ref] [bib_ref] Measurement of distribution of solids concentration on high density gas-solids flow using..., Hong [/bib_ref]. Calibration of a probe is finding a calibration curve with experiments and computations to convert the voltage time series into solids concentration time series. The appropriate calibration method must be suitable for a wide range of solids concentrations from very low concentrations up to the solids concentrations of a packed bed [bib_ref] Comparison of fibre optical measurements and discrete element simulations for the study..., Link [/bib_ref]. However, it is difficult to generate a reference suspension of particles with a given concentration, and it is nearly impossible to produce a homogeneous gas-solid flow. Thus, calibration has become a problem.
The output signals of the reflection-type probe depend, to a large extent, on the aforementioned chromaticness of the particles, and the intensity of reflected light of white and black particles of the same diameter may differ by several times, and the roughness of particle surfaces may also have a similar influence. Consequently, a calibration is required for each kind of particle. Very few studies have been carried out with regard to the effects of particle properties on the calibration curve of an optical-fiber solids concentration probe systematically and comprehensively, hindering the progress on the accurate calibration of the optical-fiber solids concentration probes.
The objective of this paper is to carry out investigation on dynamic calibration for an optical-fiber solids concentration probe in gas-solid two-phase flows. The remainder of this paper is organized as follows: a brief review of existing techniques for the calibration of optical-fiber solids concentration probes has been presented in Section 2. From the literature survey, it can be concluded that satisfactory calibration procedures are lacking in the literature and the calibration methods mainly focus on the low solids concentration. Thus, a calibration method, which is capable of calibrating the optical-fiber at nearly full range of the solids concentrations from 0 to packed bed concentration, is proposed. We combined the calibration method proposed by Hong et al. [bib_ref] Measurement of distribution of solids concentration on high density gas-solids flow using..., Hong [/bib_ref] with the calibration method used by Qi et al.. The detailed experimental setup and calibration procedures are described in Section 3. With this combined method, the probe was uniquely calibrated with two kinds of powders. Finally, the calibration results are discussed and summarized in Section 4. The effects of particle properties on the calibration curve were also investigated.
## Literature survey
Both the transmission-type probe and the reflection-type probe needed to be calibrated for their measuring range before using them for the solids concentration measurement. Several techniques for the calibration of optical-fiber probes have been developed.
Matsuno et al. [bib_ref] The use of optic fiber probes for the measurement of dilute particle..., Matsuno [/bib_ref] calibrated the optical-fiber probe containing a pair of bundles of two small plastic optical fibers by employing the free-falling particles at their terminal velocities after having travelled a certain distance. The particles were poured through a vibrating sieve at a sufficient height to fall at a uniform velocity. The particle concentration was varied by changing the weight of particle on the sieve which was located at the top of the system and also by using sieves of different apertures. The probe was set at a height sufficiently below the sieve. The solids concentration was calculated with the following equation:
[formula] s t W Au t (1) [/formula]
where ΔW is the cumulative weight of particles on the cross-sectional area A within time Δt, and u t is the calculated terminal velocity of the particles. An approximately linear relationship between the output voltage and the solids concentration was obtained. This method is usable for measuring the solids concentration certainly with a maximum concentration of 0.001. The manual vibration of the sieve and the fouling of particles will produce some erroneous data. The method is not only unsuitable to the system in which particles have high velocities but also to fast fluidized beds where the clusters of particles are formed. Cutolo et al. [bib_ref] Optoelectronic technique for the characterization of high concentration gas-solid suspension, Cutolo [/bib_ref] calibrated his probe with an apparatus consisted of a solids feed hopper and a 41 mm i.d. and 1 m high Plexiglas pipe. The particles were charged into the hopper and fell downward through a series of nets which acted as a solids distributor. The solids flow rate was regulated by the number of nets and their meshes. Measurement was made throughout the pipe core at a distance from the axis from −15 mm to +15 mm. A smaller tube with 33 mm i.d. was coaxially positioned at the bottom of the high pipe, which allowed the separation of solids falling along the walls of the larger tube. A collecting vessel placed on a balance was located below the apparatus. The average solids concentration ε s,v in the pipe of cross section A could be calculated with the following expression:
[formula] , ,() sv [/formula]
where ΔW is the weight of the particles collected during the time internal Δt, ρ is the particle density, A is the cross-section area of the pipe, U is the average falling velocity of the particles, and U(ε s,v ) is a function involved of gas pressure and composition, particle size, concentration and weight. The relationship between the output voltages and solids concentration shows good linearity for the condition that the solids concentration is below 0.1. The maximum solids concentration achieved by using this method is 0.16. Lischer and Louge [bib_ref] Optical fiber measurements of particle concentration in dense suspensions: Calibration and simulation, Lischer [/bib_ref] calibrated the optical-fiber probe used for measuring the particle volume fraction in dense suspensions against a quantitative capacitance probe. The material used was poured randomly along the probe assembly, which was mounted flush with the inside wall of a pipe that had a 15 cm i.d. They found that in a practical system, such regular calibration may be mandated by long-term variations of the average backscattered signal caused by subtle changes in the optical alignment or the quality of the fiber tip exposed to the particle suspension. However, the values from the optical-fiber probe and the capacitance probe didn't agree well.
Yamazaki et al. [bib_ref] Measurement of local solids concentration in a suspension by an optical method, Yamazaki [/bib_ref] carried out calibration experiments in a flat-bottomed cylindrical tank of 6.0 cm diameter. A known mass of solids was charged into a tank filled with water. After a steady state suspension was reached, the intensity of reflected light from the solids particles was measured by immersing the probe at various angular positions in the stirred tank. The light reflected from the solid particles in the solids-liquid mixtures had been measured in the range of solids concentration from 5% to 40% by volume. The axial concentration profiles at different radial positions agreed well, which implied that a homogeneous suspension of solids in the radial direction could be obtained by this method. It was found that measureable range for solids concentration was affected by particle species, particle diameter, particle shape, and particle color, variations in the refractive index of particles and continuous phase.
Zhou et al. [bib_ref] Voidage profiles in a circulating fluidized bed of square cross-section, Zhou [/bib_ref] calibrated their optical-fiber solids concentration probe in two liquid-solid systems. For voidages less than 0.8, the calibration was carried out in a liquid-solid fluidized bed for the reason that particles were quite uniformly distributed in such a system. The solids concentration could be obtained with the following expression:
[formula] 0 0 (1 ) s H H (3) [/formula]
where H 0 is the height of a packed bed and ε 0 is the voidage of the packed bed. Concentration data were obtained by changing the fluid velocity to vary the bed height, H. For voidages larger than 0.8, the calibration was carried out in a beaker, where a certain volume of solid particles was put into and mixed with water of known volume. The liquid-solid mixture was stirred until the particles were uniformly distributed in the water. The calibration was very nearly linear over the entire solids concentration range obtained. Different concentrations were achieved by mixing different volumes of particles into water. However, it must be pointed out that water calibration may cause measurement errors while using the probe in a gas-solid flow due to the refraction index difference between liquid and gas. The index of refraction of a liquid is greater than that of a gas, and the probes should be calibrated in the same medium for which they will be utilized [bib_ref] A novel calibration procedure for a fiber optic solids concentration probe, Zhang [/bib_ref]. Furthermore, calibrations should also be performed accordingly for each kind powders or a certain kind of powders with different diameter, different color and particle shape based on the research of Yamazaki et al. [bib_ref] Measurement of local solids concentration in a suspension by an optical method, Yamazaki [/bib_ref].
Herbert et al. [bib_ref] Application of fiber optic reflection probes to the measurement of local particle..., Herbert [/bib_ref] calibrated a single-fiber optical reflection-type probe with a technique similar to that used by Cutolo et al. [bib_ref] Optoelectronic technique for the characterization of high concentration gas-solid suspension, Cutolo [/bib_ref]. They established a stable downward flow of particles at a known velocity and in a column small enough so that a local measurement could yield a cross-sectional average value. The particles flowed from a fluidized bed feeder with a diameter of 12.5 cm, through an orifice in the center of the porous metal grid into a tube of square cross-section 8 × 8 mm, then fell 2.5 m into a collection pot. The particle flow rate from the fluidized bed could be kept stable by changing the orifice diameter. The average solids concentration could be calculated with the following expression:
[formula] , p sv pp m uA (4) [/formula]
where m p is the particle mass flow rate, u p is the particle velocity measured with an optical-fiber velocity probe, A is the tube cross-sectional area, ρ p is the particle density. During the experiments, it was found that electrostatic forces caused fine particles to cling to the probe tip and obscured light transmission, which was very undesirable since the reflected light intensity was no longer a function of the particle concentration in front of the probe, but also related to the degree of coverage of the fiber which could not be known. To overcome the electrostatic force effect, a fine-particle additive Larostat 519 was mixed with the particles.
Hong et al. [bib_ref] Measurement of distribution of solids concentration on high density gas-solids flow using..., Hong [/bib_ref] developed a method for calibrating the optical-fiber probe using a polynomial regression to correlate the output signal with the solids volume concentration in the fluidized calibration vessel. Solid particles of mass W s were placed in the fluidized vessel made of transparent glass and then fluidized to the full vessel volume after closing the top end of the vessel with a filter screen. The overall volume concentration of solids within the full vessel could be calculated with the following expression:
[formula] s s s W AH (5) [/formula]
where ρ s is the solids density, A is the cross-sectional area of the vessel and H is the vessel height. They found that simply assuming a linear relationship between the output signal and solids concentration may cause a significant deviation from experiment in the range of concentration above 0.05. They also pointed out that a powder with very homogeneous fluidized behavior would have a more reliable calibration curve using their method. The disadvantage of this method was that the formation of bubbles in the bed during calibration which would cause some scatter of the calibrating points. San José et al. [bib_ref] Local bed voidage in conical spouted beds, San [/bib_ref] calibrated the optical-fiber probe used for voidage measurement of a conical spouted bed. For the spout zone, the calibration was carried out in a 60 mm i.d. column, where the probe had been introduced at a given level. The solid was fed by a hopper to the column. A linear relationship between the intensity of the reflected light and the volume fraction of the bed occupied by the particles, 1 − ε, was obtained. Bed voidage, ε, which was changed by adjusting the solid mass flow rate in the feed, Q, could be calculated with the following expression:
[formula] (1 ) z Q Av (6) [/formula]
where ρ is the particle density, A is the column cross sectional area, v z is the velocity of the particles along the longitudinal direction at the probe zone, which can be calculated as follows:
[formula] e z d v (7) [/formula]
where d e is the effective distance between the two receiving fibers, τ is the delay time between the two signals. For the annular zone, the calibration consisted of measurement in moving beds, in a column of 60 mm i.d., using different particle sizes. Three kinds of different experiments were carried out: packed beds, beds loosened to the maximum and partially loosened beds. They found that the position of the probe in the bed did not affect the resulting calibration curve. This method has two disadvantages that the solids concentration along the cross-section of the downer column is mal-distribution and using the measurement of particle velocity with optical-fiber to evaluate solids concentration is not absolutely a good measurement technique.
Zhang et al. [bib_ref] A novel calibration procedure for a fiber optic solids concentration probe, Zhang [/bib_ref] proposed a back pressure control method to calibrate a multi-fiber optical reflection probe in a downer to obtain quantitatively precise solids concentration. The calibration apparatus was similar to that of Herbert et al. [bib_ref] Application of fiber optic reflection probes to the measurement of local particle..., Herbert [/bib_ref] , and the significant improvement was adding a back pressure control system by sealing the bottom collection vessel. By using quick closing valves [bib_ref] Measurements of solids concentration and axial solids velocity in gas-solid two-phase flows, Nieuwland [/bib_ref] , the particles were trapped to determine various solids concentrations to compare with the data obtained with optical-fiber probe. The solids concentration could be obtained as large as 0.56 because the increase of the back pressure decreased the particle velocity and increased the solids concentration. Meanwhile, an iteration procedure was employed to modify the initial calibration curves. They also found that the probes were sensitive to minor variations of particle color and reflective properties. However, the flow was not entirely radial uniform even by using a vibrator to distribute the particles in the downer, which would cause calibration errors.
Johnsson et al. [bib_ref] Measurements of local solids volume-fraction in fluidized bed boilers, Johnsson [/bib_ref] carried out calibration experiments of an optical-fiber probe in a cold CFB riser, and compared the output signals with those obtained by an optical reference probe, which was calibrated with a guarded capacitance probe. The guarded capacitance probe was calibrated from measurements in a packed bed. They found that the reference probe and the optical-fiber probe gave similar response in amplitude and frequency with respect to variations in solids volume-fraction. The shape of the calibration function of the reference probe was also valid for the optical-fiber probe. The calibration and the reference measurement were carried out at ambient conditions, while the measurements were done at 850 °C with the assumption that the shape of the calibration function obtained under ambient condition was valid at elevated temperatures. They believed that it was a reasonable assumption because the shape of the calibration function depended on the optical properties of the probe and not on the temperature of the gas and particles.
Cui et al. [bib_ref] Comparison of Measurement Techniques of Local Particle Concentration for Gas-Solid Fluidization, Cui [/bib_ref] suggested a novel calibration method and correlation for different optical-fiber probes. They made a series of uniform mixture samples of FCC and amorphous transparent polystyrene by blending the two materials in a Brabender mixer at 200 °C , shaped the samples into a cubic form in a size of 20 mm, and finally polished them to attain good optical properties. For different samples, the FCC particle concentration varied from 0 to that of the minimum fluidization state. The mixtures were used to simulate gas-solid flow systems with different solids concentrations, in which the transparent polystyrene was seen as air. The transparent polystyrene had refractive index higher than that of air. They found that the cross probe and parallel probes with glass window had similar calibration curves. The output voltages of the probe increased sharply with increasing solids concentration at low concentration but increased slowly at high solids concentration. This method is very interesting and novel. The probes were calibrated with a homogeneous dispersion of solids in the polymer-solid cubes with different and exact solids concentrations which is representative of different solids concentrations inside the gas-solid fluidized bed.
Rundqvist et al. [bib_ref] Dual optical fibre measurements of the particle concentration in gas/solid flows, Rundqvist [/bib_ref] proposed an improvement in the design of a dual optical-fiber probe as well as a general calibration theory. They calibrated the probes in a small stirred tank filled with water, and a controlled mass of particles was added. The results of their calibration experiments showed some discrepancies relative to theoretical calibrations. A disproportional fraction of the particles were observed to reside close to the walls and the bottom of the calibrating vessel, which hindered the calibration experiments, leading to lower volume fractions than expected and offered a possible explanation to the above mentioned discrepancies. It was concluded that the calibration theory scaled well with probe size, particle size and particle volume fraction, although the exact shape of the calibration function could not be verified exactly.
Liu et al. [bib_ref] Novel multifunctional optical-fiber probe: I. Development and validation, Liu [/bib_ref] [bib_ref] Novel multifunctional optical-fiber probe: II. High-density CFB measurements, Liu [/bib_ref] calibrated three-fiber optical probes with two units. The first was a dropping/trapping technique. Particles fall into a collection vessel from an incipiently fluidized bed through a short tube located in the center of a punched plate distributor covered with fine wire mesh into a 12 mm i.d. tube. After a steady flow was obtained, two slide plates located 32 mm apart were closed quickly and simultaneously to trap the particles in a section of pipe where the probe inserted. By using different sizes of feeding tube and the flow rate of aeration air, different solids concentration could be obtained. The second was to obtain a water-FCC suspension in a well-stirred beaker. The probe with and without the quartz glass window was calibrated using both the two calibration units, and the results were compared with simulation predictions. The simulation results were in good agreement with the calibration results. They found that without the glass window, the calibration curves were highly nonlinear, which meant that the protective window could improve the linearity of the calibration curve.
They also suggested that one should not directly apply calibrations obtained in liquids to calibrate probes for gas-solids systems.
Magnusson et al. [bib_ref] Dual fibre optical probe measurements of solids volume fraction in a circulating..., Magnusson [/bib_ref] calibrated a dual fiber-optical probe based on the calibration theory proposed by Rundqvist et al. [bib_ref] Dual optical fibre measurements of the particle concentration in gas/solid flows, Rundqvist [/bib_ref] in a circulating fluidized bed, for the reason that circulating fluidized beds provide a wide range of flow conditions and solids volume fractions. The particle volume fractions measured by the optical-fiber probe were compared to the pressure drop measured for a range of operation conditions. The relation between the pressure drop and the solids volume fraction could be expressed as follows:
[formula] () s s g acc P gz p (8) [/formula]
where ε s is the solids volume fraction, ρ s and ρ g are the solids and gas phase densities, g is the acceleration due to gravity and z is the distance between the pressure measuring taps, p acc is the pressure drop due to acceleration of the particles, which may be neglected. They found that glare points on the particles and the beam length from the probe to the particle determines the curvature of the calibration curve. The values obtained with pressure drop measurements and the values of optical-fiber probe measurements showed poor agreement. The measurement volumes for the optical probe and the pressure drop are different from each other. Thus, application of this method for calibration of optical-fiber solids concentration probe is not satisfactory.
As can be seen from the literature review above, most of the researchers performed their calibration procedure and obtained specific kind of calibration curve, either linear or non-linear. All the calibration procedures were established by comparing the optical-fiber probe output signals to the concentration values obtained by the traditional methods for direct measurement of solids concentration, which could be mainly divided into three categories. The first is to use quick closing valves [bib_ref] Measurements of solids concentration and axial solids velocity in gas-solid two-phase flows, Nieuwland [/bib_ref] , with which the column to be studied can be positioned in sections of suitable length. The two valves were used to trap the solids within the desired section of the downer. After a certain period of time, the valves were closed simultaneously. The solids contained in the section can be collected and weighed, thus the solids concentration may be determined. This method was used by Zhang et al. [bib_ref] A novel calibration procedure for a fiber optic solids concentration probe, Zhang [/bib_ref] and Issangya et al.. The second is to measure the pressure drop over a certain section of a riser tube, and the Bernoulli equation neglecting wall friction and acceleration forces was solved for solids concentration. This method was adopted by Magnusson et al. [bib_ref] Dual fibre optical probe measurements of solids volume fraction in a circulating..., Magnusson [/bib_ref]. The third is to build a stable downward flow system with particle density deduced from mass flux of particles and measurement where phase velocities were nearly equal. This method was widely used by researchers with appropriate modifications, like Matsuno et al. [bib_ref] The use of optic fiber probes for the measurement of dilute particle..., Matsuno [/bib_ref] , Cutolo et al. [bib_ref] Optoelectronic technique for the characterization of high concentration gas-solid suspension, Cutolo [/bib_ref] , Lischer and Louge [bib_ref] Optical fiber measurements of particle concentration in dense suspensions: Calibration and simulation, Lischer [/bib_ref] , Herbert et al. [bib_ref] Application of fiber optic reflection probes to the measurement of local particle..., Herbert [/bib_ref] and so on. In addition, liquid fluidized beds had been used by some investigators with the purpose to overcome the difficulty of obtaining stable suspensions of solids in gases, like Yamazaki et al. [bib_ref] Measurement of local solids concentration in a suspension by an optical method, Yamazaki [/bib_ref] , Zhou et al. [bib_ref] Voidage profiles in a circulating fluidized bed of square cross-section, Zhou [/bib_ref] , Rundqvist et al. [bib_ref] Dual optical fibre measurements of the particle concentration in gas/solid flows, Rundqvist [/bib_ref] , and Liu et al. [bib_ref] Novel multifunctional optical-fiber probe: I. Development and validation, Liu [/bib_ref] [bib_ref] Novel multifunctional optical-fiber probe: II. High-density CFB measurements, Liu [/bib_ref]. However, it is worthwhile to mention that the validity of such a calibration in gaseous suspensions is questionable due to the differences in the refractive index between gases and liquids [bib_ref] Calculation of optic fibres calibration curves for the measurement of solids volume..., Amos [/bib_ref] , and the probes should be calibrated in the same medium for which they will be utilized. Different calibration methods used by researchers are summarized in .
The accuracy of solids concentration measurement by using an optical-fiber solids concentration probe is strongly dependent on the precision of the calibration technique utilized. The calibration of optical-fiber solids concentration probe in gas-solid environment is challenging due to the heterogeneity and instability of gas-solids flow. It is difficult to offer a series of standard gas-solid flow covering solids concentrations from 0 to packed bed. Because it is rather difficult to maintain a homogeneous gas-solids flow at high solids concentrations, the majority of calibration methods developed to date are focused on obtaining relatively homogeneous gas-solids flow at low solids concentrations. When applying them to the practical measurements, the calibration methods may be inaccurate or problematic. There are no widely-accepted calibration methods which cover a wide range of solids concentrations up to now. Thus, a feasible and simple method for the calibration of optical-fiber solids concentration probe which could solve the above mentioned questions of the existing calibration methods is urgently needed to be developed. More research on the calibration of optical-fiber solids concentration probe, especially experimental, is still required.
## Experiment setup and calibration procedure
This study utilized a model PC6M optical-fiber solids concentration probe which was developed by the Institute of Process Engineering, Chinese Academy of Science, Beijing, China. This measurement system was composed of PC6M concentration measurement main unit, optical-fiber probes, and the signal cable, A/D converter and application software. The probe tip is 4 mm in diameter and contains approximately 8,000 emitting and receiving quartz fibers, each of a diameter of about 25 μm. These fibers are arranged in an alternating array, corresponding to emitting and receiving layers of fibers. The active area, where the fibers are located, is approximately 2 mm × 2 mm. The fiber tips are protected from the material by a glass window with thickness of 0.2 mm. The received light reflected by the particles is multiplied by the photo-multiplier and converted into a voltage signal. The voltage signal is further amplified and fed into a computer. The high voltage adjustment is used to adjust the upper measuring limit or full scale of the instrument. There is a zero voltage potentiometer which adjusts the output signal to zero when no powder is on the tip of the probe, thus the offsets of PC6M were set at zero with empty black box and the gains roughly at 4.5 V with packed box (less than the full range of 5 V), making the calibration procedure respond to most of possible particle concentrations. In order to make the day-to-day measurement comparable, the lamp voltage which sets the power voltage of the light source, and the gain factor should be kept constant during the calibration procedures [bib_ref] A novel calibration procedure for a fiber optic solids concentration probe, Zhang [/bib_ref]. [fig_ref] Figure 3: The optical-fiber probe system[16] [/fig_ref] shows the schematic diagram of the optical-fiber probe system.
## Receiving fibers
A Plexiglas column was used to calibrate the probes. As shown in [fig_ref] Figure 4: Calibration [/fig_ref] , the column is 50 mm in diameter and 300 mm high. A perforated Plexiglas plate covered with fine screens was employed as the gas distributor. Optical-fiber probes were installed along the column with their tips located on the axis of the column. Two pressure taps, one at the wind room and the other at the exit section of the column were provided to record the pressure drops. Pressure drop was measured by a 1 m long U-tube manometer, and water was used as the manometric fluid. The fluidizing gas was supplied by a nitrogen cylinder to the wind room which is 50 mm in diameter and 100 mm in length below the gas distributor. Bubble suppressor was installed in the calibration column, which contained a set of metal meshes with 25 mm axial pitches, the diameter of the metal mesh was 45 mm and the opening of each metal mesh was 3 mm × 3 mm with 0.5 mm steel wire. The top end of the column was covered with a filter screen to prevent the fine particles escaping from the column. Meanwhile, all metal portions of the whole calibration apparatus were carefully connected and grounded to eliminate static electricity effect. Glass beads and quartz sand were used to investigate the particle properties. Both of the two powders were sieved into three narrow distribution parts, and their physical properties are listed in [fig_ref] Table 2: Physical properties of powders [/fig_ref]. The particle size distributions were measured with a laser particle analyzer (LS, Beckman Coulter Inc., Brea, CA, USA), as shown in [fig_ref] Figure 5: Particle calibration process, the optical-fiber probe was inserted into the column center... [/fig_ref]. Because the particle diameter is much smaller than the bundle diameter, light is reflected by the particles in the measurement volume, allowing the probe to detect to measure the solids concentration. The novel calibration procedure in the present study are consisted of two parts: one for high solids concentration (Method I) and the other for low solids concentration (Method II). The calibration procedure for high solids concentrations was similar to that of Qiused a Pseudo Bubble-Free Fluidized Bed (PBFF). The powder was filled into the column with an initial bed height of about 100 mm, then fluidized by the fluidizing gas and was better distributed throughout the column with the assistance of bubble suppressors compared to without the installing of bubble suppressors.
From the results obtained by this paper, this kind of calibration procedure was limited to solids concentration higher than 0.2. Thus, for the low solids concentration, the calibration data was obtained with the method similar to Hong et al. [bib_ref] Measurement of distribution of solids concentration on high density gas-solids flow using..., Hong [/bib_ref] and the detail procedures were with reference to Wang et al. [bib_ref] Experimental study on the solid velocity in horizontal dilute phase pneumatic conveying..., Wei [/bib_ref]. All the calibration experiments were repeated 6 times under the same condition for the accuracy of the reproducibility. To minimize the influence of fine particles adhering to the tip surface of the probe on the calibration, particularly at the relatively high concentration, the optical-fiber probe was removed for cleaning of its surface for every test. In order to avoid the effect of any outside light source, the calibration column was covered with black cloth after fluidization of the solids particles. The sampling rate of the optical-fiber probe was 1 kHz and the sampling time approximately 4 s.
# Results and discussion
Due to that the complexity of the heterogeneous gas-solids flow manifests in the irregular, non-periodic variation of solids concentration with time, the typical time-resolved signals from the probe exhibit sharp spikes that correspond to the passage of individual particles in the near vicinity of the probe [bib_ref] Optical fiber measurements of particle concentration in dense suspensions: Calibration and simulation, Lischer [/bib_ref]. Different from pressure fluctuation signals, the solids concentration signals show binary behavior [bib_ref] Analysis of the chaotic dynamics of a high-flux CFB riser using solids..., Manyele [/bib_ref]. The peaks of the signals stand for high solids concentration and the valleys represent low solids concentration. During the calibration experiments, after reaching steady state, the mean output voltages of the probe were measured by immersing the probe at four radical positions: r/R = 0, r/R = 0.2, r/R = 0.4, r/R = 0.8. The radial voltage values of the probe are plotted in [fig_ref] Figure 6: The probe output voltages of GB 3# at four different radial positions [/fig_ref]. It can be seen that the output voltages of the probe at different radial positions basically agree well. The output voltages of the probe recorded near the wall is higher than those of the other three radial positions, which indicates higher solids concentration near the wall due to the wall effect. It can be inferred that relatively homogeneous suspension of solids was obtained. In order to ensure that day-to-day measurements could be compared on an equal basis and to eliminate any parasitic effects caused by this particular apparatus which may prevent the use of the calibration curve in other systems, the output voltages were normalized according to the following expression [bib_ref] Application of fiber optic reflection probes to the measurement of local particle..., Herbert [/bib_ref] :
[formula] 0 0 p VV v VV (10) [/formula]
where V 0 is the output voltage when the solids concentration is zero, and V p is the output voltage when the solids concentration is equal to packed bed concentration. According to the research of Cui et al. [bib_ref] Comparison of Measurement Techniques of Local Particle Concentration for Gas-Solid Fluidization, Cui [/bib_ref] , the calibration correlations of the normalized output voltages and normalized solids concentration can be expressed as follows:
[formula] , (1 ) s sp av av (11) [/formula]
where ε s,p is the packed bed solids concentration, ε s,p/ ε s,p is normalized solids concentration, and a is the calibration coefficient. The relationships between normalized output voltages and normalized solids concentration are shown in [fig_ref] Figure 7: Calibration curves for Glass beads [/fig_ref] , it can be seen that the output voltages increases sharply with increasing solids concentration at low solids concentrations, while increases slowly at high solids concentrations, which is consistent with the variation trend obtained by Cui et al. [bib_ref] Comparison of Measurement Techniques of Local Particle Concentration for Gas-Solid Fluidization, Cui [/bib_ref]. It may due to that for higher solids concentrations, the probe has a smaller measuring volume and lead to a more 'local' measurement because particles in the measuring volume woule be blocked by particles in front of them. Meanwhile, it can also been seen that the calibration coefficient a increases with the increasing of particle size. Method II was used to calibrate the probe with a solids concentration range from 0.02 to 0.2 for both of the two powders. Method I worked at the solids concentrations from 0.26 to 0.53 for glass beads and from 0.22 to 0.5 for quartz sand. [fig_ref] Figure 7: Calibration curves for Glass beads [/fig_ref] , it can also be seen that method I and method II together are capable of calibrating the optical-fiber probe at nearly full range of the solids concentrations from 0 to packed bed concentration. The suspensions encountered in practical problems of chemical engineering usually consist of particles with statistically varying irregular shapes, size and surface properties, locations and spatial orientations which are altogether unknown and for practical problems also may not be exactly describable [bib_ref] Estimation of the effective measuring volume of single-fibre reflection probes for solid..., Rensner [/bib_ref]. The effect of particle size on calibration results is shown in [fig_ref] Figure 9: Effect of particle size on the calibration results [/fig_ref]. It can be inferred that the output voltage of the optical fiber probe has a tendency to increase with decreasing particle size for the whole range of solid concentrations, but the increase tendency is not so obvious when the solids concentration less than 0.1. The intensity of the backscattered light is a function of the solids concentration and the mean particle size [bib_ref] Measurement of local solids concentration in a suspension by an optical method, Yamazaki [/bib_ref] [bib_ref] Light backscattering as a technique to measure solids particle size and concentration..., Bos [/bib_ref]. Yamazaki et al. [bib_ref] Measurement of local solids concentration in a suspension by an optical method, Yamazaki [/bib_ref] found that the intensity of the reflected light is affected by the particle diameter and the back-scattered light reach the probe decreases as the particle diameter increases. This may be attributed to the increase in the average path length of the light beam, which is caused by an increased particle diameter. Bos et al. [bib_ref] Light backscattering as a technique to measure solids particle size and concentration..., Bos [/bib_ref] indicated that the intensity of the reflected light increases as the particle diameter decreases and the solids concentration increases. The present finding is consistent with their results. Meanwhile, Qialso obtained similar conclusions that the output voltage decreases substantially with increasing particle size. According to Amos et al., the relationship between the probe output and the solids concentration and the variation of this relationship seemed to be strongly dependent on the relationship between the particle size and the fiber diameter. With fixed fiber diameter, the relationship between the probe output and the solids concentration seemed to have the same slop at the low solids concentrations. In our study, the offset of the probe were adjusted to nearly zero with empty black box for each power with different particle diameter. Thus, the difference of calibration curves within the low solids concentration range between different particle sizes is not obvious. Amos et al.also found that particle size effect was greater at higher solids concentrations, and that increasing the particle diameter while keeping constant the solids concentration and fiber diameter would lead to more light penetrate into the solid suspension. The light penetrating into the bed deeper than one particle would never be reflected out of the solids suspension. Meanwhile, the discrepancy of calibration results with different particle size exists at moderate solids concentration may also have some relationship with some heterogeneity effect in the gas-solid system due to that the inherent fluctuations are unable to be completely avoided in a gas-solid flow system. During the calibration experiments, it was found that the bubbles are more likely to form and expand at moderate solids concentrations. The bubble suppressor can reduce the generation of bubbles and avoid inherent fluctuations inside the calibration column to a certain extent, but it can't suppress the generation of bubbles totally. The degree of heterogeneity was relatively larger at moderate solids concentrations. Thus, a powder with very homogeneous fluidized behavior would be expected to have a more reliable calibration curve with this applied methodology. GB 1# and QS 1# are similar in particle size, but different in sphericity and color. Thus, comparisons are difficult to be made directly between these two powders, as shown in [fig_ref] Figure 1: Two different types of optic fiber probes[19] [/fig_ref]. A batch of fresh glass beads with the same particle size as GB 1# but has a color of white, which was named GB 0#, was used to compare with the above mentioned two powders separately. The effect of particle color on calibration results is shown in [fig_ref] Figure 1: Two different types of optic fiber probes[19] [/fig_ref]. The difference between GB 0# and GB 1# was color, and GB 1# was darker than GB 0#. It can be seen from [fig_ref] Figure 1: Two different types of optic fiber probes[19] [/fig_ref] that GB 0# has higher output voltages than those of GB 1#, which is due to that dark powder reflects less light [bib_ref] A novel calibration procedure for a fiber optic solids concentration probe, Zhang [/bib_ref]. The effect of particle sphericity on calibration results is shown in [fig_ref] Figure 1: Two different types of optic fiber probes[19] [/fig_ref]. The calibration curves of GB 0# and QS 1# overlapped together. For sand particles, the sphericity is in the region of 0.8~0.98 [bib_ref] Dual optical fibre measurements of the particle concentration in gas/solid flows, Rundqvist [/bib_ref] which is usually less than that of glass beads. Rundqvist et al. [bib_ref] Dual optical fibre measurements of the particle concentration in gas/solid flows, Rundqvist [/bib_ref] pointed out that the sampling light scattered from irregular particles will approximate the light scattered from the same number of spherical particles if a sufficient number of particles are considered, as the particles are oriented randomly in the suspension. The irregular object will appear more spherical as angular velocity increases, which is equivalent to sampling the same object from different angles. The impact of sphericity on the calibration theory was neglected in their research. Thus, it can be inferred that the effect of particle color on calibration result is more predominant than that of sphericity, Qialso obtained the same conclusion. The calibration curves in [fig_ref] Figure 1: Two different types of optic fiber probes[19] [/fig_ref] are very close and overlapped at solids concentrations less than 0.2, which means that the calibration curve is less sensitive to the particle color when the solids concentration is low. The difference of the calibration curve at solids concentration more than 0.2 may due to the color difference. The GB 1# are darker in color than QS 1#. Grey glass beads have the tendency to absorb more light compared to white quartz sand. It can be seen that under the same solids concentration, the output voltages of QS 1# are higher than that of GB 1#.
# Conclusions
Different calibration methods of optical-fiber solids concentration probes reported in the literature were reviewed in this paper. Satisfactory calibration procedures are lacking in the literature, and the exact shape of the calibration function has not been verified exactly. A combined calibration method, which is capable of calibrating the optical-fiber probe at nearly full range of the solids concentrations from 0 to packed bed concentration, is proposed. With this combined method, the probe was uniquely calibrated with two kinds of powders. The effects of particle properties (particle size, sphericity and color) on the calibration results were comprehensively investigated. From the experiments carried out here, it can be concluded that the output voltage has a tendency to increase with the decreasing particle size, and the effect of particle color on the calibration curve is more predominant than that of sphericity.
[fig] Figure 1: Two different types of optic fiber probes[19] (a) transmission-type; [/fig]
[fig] Figure 3: The optical-fiber probe system[16]. [/fig]
[fig] Figure 4: Calibration [/fig]
[fig] Figure 5: Particle calibration process, the optical-fiber probe was inserted into the column center between two successive bubble suppressor plates. The experiments were carried out under different ratio of H/H mf , where H mf is the bed height at minimum fluidization state and H is the bed height under other fluidizing gas flow rate. The solids concentration could be calculated according to the following equation: [/fig]
[fig] Figure 6: The probe output voltages of GB 3# at four different radial positions. [/fig]
[fig] Figure 7: Calibration curves for Glass beads: (a) GB 1#; (b) GB 2#; (c) GB 3#. [/fig]
[fig] Figure 8: Calibration [/fig]
[fig] Figure 9: Effect of particle size on the calibration results: (a) Glass beads; [/fig]
[table] Table 1 Table 1: Different calibration methods. Cont. [/table]
[table] Table 2: Physical properties of powders. [/table]
|
Glycosyltransferase GLT8D2 Positively Regulates ApoB100 Protein Expression in Hepatocytes
Non-alcoholic fatty liver disease (NAFLD) is characterized by triglyceride (TG) accumulation in hepatocytes. Very low density lipoprotein (VLDL) is a major secretory product of the liver that transports endogenously synthesized TG. Disrupted VLDL secretion may contribute to the accumulation of TG in hepatocytes. ApoB100 (apolipoprotein B100) is a glycoprotein and an essential protein component of VLDL. Its glycosylation may affect VLDL assembly and secretion. However, which glycosyltransferase catalyzes apoB100 glycosylation is unknown. In this study, we cloned the GLT8D2 (glycosyltransferase 8 domain containing 2) gene from HepG2 cells and generated a series of plasmids for in vitro studies of its molecular functions. We discovered that GLT8D2 was localized in the ER, interacted with apoB100, and positively regulated the levels of apoB100 protein in HepG2 cells. Based on these results, we propose that GLT8D2 is a glycosyltransferase of apoB100 that regulates apoB100 levels in hepatocytes.
# Introduction
Non-alcoholic fatty liver disease (NAFLD) has become the most common chronic liver disease in the United States and other western countries [bib_ref] Protective effect of probucol on liver injury induced by carbon tetrachloride in..., Zhan [/bib_ref] [bib_ref] Roles of liver innate immune cells in nonalcoholic fatty liver disease, Zhan [/bib_ref]. It is characterized by triglyceride (TG) accumulation in hepatocytes [bib_ref] Update on the treatments of non-alcoholic fatty liver disease (NAFLD), Quercioli [/bib_ref]. However, the mechanism of this triglyceride accumulation remains elusive. Very low density lipoprotein (VLDL) is a major secretory product of the liver that transports endogenously synthesized lipids, mainly TG [bib_ref] The degradation of apolipoprotein B100: Multiple opportunities to regulate VLDL triglyceride production..., Fisher [/bib_ref]. It was recently reported that the secretion rate of VLDL increased linearly with increasing intra-hepatic TG disposal, while that in NAFLD patients did not [bib_ref] Alterations in adipose tissue and hepatic lipid kinetics in obese men and..., Fabbrini [/bib_ref]. Therefore, disrupted VLDL secretion may contribute to the accumulation of triglyceride in hepatocytes. VLDL consists of cholesterol, phospholipids, triglycerides and apolipoproteins that include apoB100, apoC, and apoE [bib_ref] A treasure trove for lipoprotein biology, Lusis [/bib_ref]. Based on the complexity of VLDL, We speculated that abnormalities in VLDL assembly may be the main cause of VLDL secretion disruption in NAFLD patients.
ApoB100, the essential protein component of VLDL [bib_ref] Overexpression of the tumor autocrine motility factor receptor Gp78, a ubiquitin protein..., Liang [/bib_ref] , is a glycoprotein. Glycosylation aids the folding of nascent polypeptide chains and stabilizes mature glycoprotein's conformation [bib_ref] Overexpression of the tumor autocrine motility factor receptor Gp78, a ubiquitin protein..., Liang [/bib_ref]. Therefore, glycosylation may affect apoB100's folding and maturation, which further affects VLDL assembly and secretion. Glycosylation is catalyzed by glycosyltransferase enzymes, and over 160 human glycosyltransferase genes have been cloned [bib_ref] Phylogenomic analysis of UDP glycosyltransferase 1 multigene family in Linum usitatissimum identified..., Barvkar [/bib_ref] [bib_ref] Human disease glycomics/proteome initiative workshop and the 4th HUPO annual congress, Taniguchi [/bib_ref]. However, which glycosyltransferase catalyzes apoB100 glycosylation is unknown. Recently, we cloned a new glycosyltransferase gene, namely GLT8D2. In the present study, we observed that GLT8D2 interacted with apoB100, and this interaction affected apoB100 protein expression in hepatoma cell line (HepG2 cells). These results suggested that GLT8D2 may be a glycosyltransferase of apoB100.
# Results
## Amplification of glt8d2 gene
The GLT8D2 gene is amplified by using the GLT8D2 cDNA as template and two specific primers. As shown in [fig_ref] Figure 1: Electrophoresis of GLT8D2 PCR product on 1% agarose gel [/fig_ref] , the GLT8D2 PCR product was approximately 1050 bp, as expected.
## Successful construction of plasmids
The amplified and purified GLT8D2 gene fragment was ligated into the pEGFP-C1 vector to construct the pEGFP-C1-GLT8D2 plasmid. The integrity of recombinant vector pEGFP-C1-GLT8D2 was confirmed by double restriction enzyme digestion with KpnI and XhoI. As shown in [fig_ref] Figure 2: Electrophoresis of recombinant GLT8D2 on 1% agarose gel [/fig_ref] , two expected bands, one at 4700 bp and another at 1050 bp, were observed after the digestion. Similarly, we constructed the plasmid of pEGFP-C1-mut-GLT8D2 by ligating the purified mutant GLT8D2 gene fragment into the pEGFP-C1 vector, which was also confirmed by restriction enzyme digestion and electrophoresis (data not shown). The success of plasmid construction was further confirmed by sequencing the inserted gene fragments in pEGFP-C1-GLT8D2 and pEGFP-C1-muta-GLT8D2.
## Knockdown efficiency of pcdna6.2-glt8d2 shrnas in hepg2 cells
The pcDNA6.2-GLT8D2 shRNA-1-4, pcDNA6.2-GLT8D2 shRNA-IR, and pcDNA6.2-negative were transfected into HepG2 cells. As shown in [fig_ref] Figure 3: Knockdown efficiency of GLT8D2 shRNAs in HepG2 cells revealed by qRT-PCR [/fig_ref] , only pcDNA6.2-GLT8D2 shRNA-1 and -4 led to significant reduction in the mRNA expression of GLT8D2 (p < 0.001 and 0.01 versus control, respectively), with pcDNA6.2-GLT8D2 shRNA-1 showing about 94% knockdown efficiency. pcDNA6.2-GLT8D2 shRNA-3 did not affect the mRNA expression of GLT8D2 (p > 0.05 versus control). pcDNA6.2-GLT8D2 shRNA-2 and pcDNA6.2-GLT8D2 shRNA-IR led to significant increases in the mRNA expression of GLT8D2 (both p < 0.05).
## Intracellular localization of the glt8d2 protein
The plasmids of pEGFP-C1-GLT8D2 and pDsRED-N1-calreticulin were co-transfected into HepG2 cells, and the expressed GLT8D2 and calreticulin were detected by confocal microscopy. The EGFP-C1-GLT8D2 fusion protein emits green fluorescence, the DsRED-N1-calreticulin fusion protein emits red fluorescence, and the DAPI emits blue fluorescence. Double-fluorescence analysis showed that the expression of GLT8D2 (green) in HepG2 cells overlapped with calreticulin (red) [fig_ref] Figure 4: Co-localization of GLT8D2 and calreticulin in HepG2 cells [/fig_ref] , indicating that GLT8D2 protein is localized in the ER around the nucleus in HepG2 cells.
## Direct interaction of glt8d2 and apo-b100
To determine whether GLT8D2 binds to apoB100, co-IP was carried out with HepG2 cell lysate and specific antibodies against GLT8D2 and apoB100. As shown in , the GLT8D2 protein was co-immunoprecipitated with the apoB100 protein by the anti-apoB100 antibody, suggesting that GLT8D2 and apoB100 form a complex in HepG2 cells. We observed the molecular weight of GLT8D2 that was consistent with its expected size of 55 kDa. . GLT8D2 may be interacted with apo-B100 in HepG2 cells. The result showed co-immunoprecipitation of apo-B100 and GLT8D2 in extracts of HepG2 cells. Anti-GLT8D2 immunoprecipitation (IP) followed by anti-apo-B100 Western blotting (WB) or anti-apo-B100 IP followed by anti-GLT8D2 Western blot. The rabbit irrelevant IgG was used as immunoprecipitated control.
## Glt8d2 affects apob100 expression in hepg2 cells
To determine the effects of GLT8D2 on the expression of apoB100, we over-expressed GLT8D2 by transfecting the GLT8D2 expression plasmid, pEGFP-C1-GLT8D2, and knocked-down the expression of GLT8D2 by transfecting pcDNA6.2-GLT8D2 shRNA-1 in HepG2 cells. We also introduced the mutant GLT8D2 into HepG2 cells by transfecting the GLT8D2 mutant plasmid, pEGFP-C1-mut-GLT8D2. As shown in [fig_ref] Figure 6: Western blot analysis of apoB100 expression in HepG2 cells with changed expression... [/fig_ref] , the expression of apoB100 protein was increased by GLT8D2 overexpression in HepG2 cells; the apoB100 level was also reduced with silenced expression of GLT8D2. Expression of the mutant GLT8D2 in HepG2 cells led to down-regulated expression of apoB100. These results suggested that the expression level of GLT8D2 is positively correlated with that of apoB100 in HepG2 cells.
# Discussion
As a member of the glycosyltransferase 8 family, GLT8D2 is a 349 amino acid single-pass type II membrane protein encoded by a gene that is located on human chromosome 12q23.3. The first six amino acid residues extend to the cytoplasm; the No. 7-24 amino acid residues are in the plasma membrane; the amino acid residues No. 25-349 are in luminal compartments. GLT8D2 is also a glycoprotein with only one glycosylation site, Asn 234. In this study, we successfully amplified the GLT8D2 gene and constructed a series of GLT8D2 plasmids, providing valuable materials for future studies on GLT8D2.
The endoplasmic reticulum (ER) plays an essential role in the folding and processing of newly synthesized secretory membrane proteins, which is strictly calcium-dependent [bib_ref] Disturbances of the functioning of endoplasmic reticulum: A key mechanism underlying neuronal..., Paschen [/bib_ref]. The ER is also crucial for glycoprotein glycosylation [bib_ref] The endoplasmic reticulum as the extracellular space inside the cell: Role in..., Csala [/bib_ref]. In this study, we observed the localization of GLT8D2 glycosyltransferase in the ER, consistent with its expected role in glycoprotein glycosylation.
ApoB100 is a large secretory glycoprotein with 4536 amino acid residues, and its molecular weight has been calculated to be 513 kDa [bib_ref] Site-specific glycosylation analysis of human apolipoprotein B100 using LC/ESI MS/MS, Harazono [/bib_ref]. However, the glycosyltransferase that catalyzes apoB100 glycosylation is still unknown. It was suggested by Ihara et al. that N-Acetylglucosaminyltransferase III may be related to apoB100's glycosylation in hepatocytes [bib_ref] Ectopic expression of N-acetylglucosaminyltransferase III in transgenic hepatocytes disrupts apolipoprotein B secretion..., Ihara [/bib_ref]. However, no direct evidence for the interaction between N-Acetylglucosaminyltransferase and apoB100 was provided in that study. By using co-immunoprecipitation, we found that GLT8D2 protein was bound to apoB100, suggesting that GLT8D2 may be a glycosyltransferase of apoB100.
ApoB100 is synthesized in hepatocytes. Newly synthesized apoB100 is translocated across the rough endoplasmic reticulum membrane into the lumen, where apoB100 is in at least two pools: a heavy pool, most of which is degraded in situ, and a lighter pool, which moves from the rough endoplasmic reticulum lumen through the secretory compartments to the trans-Golgi; it is then packaged with lipid and secreted as VLDL [bib_ref] Intracellular events in the assembly of very-low-density-lipoprotein lipids with apolipoprotein B in..., Cartwright [/bib_ref]. As a result, degradation and secretion play an important role in the quality control of the apoB100 protein in hepatocytes. It has been shown that 30%-75% of newly synthesized apoB100 is degraded rapidly within 2-3 h [bib_ref] Apolipoprotein B100 biogenesis: A complex array of intracellular mechanisms regulating folding, stability,..., Rutledge [/bib_ref] [bib_ref] Tunicamycin induces ubiquitination and degradation of apolipoprotein B in HepG2 cells, Liao [/bib_ref]. The expression of apoB100 is also regulated post-transcriptionally [bib_ref] Conformational changes in apolipoprotein B modulate intracellular assembly and degradation of ApoB-containing..., Macri [/bib_ref]. Glycosylation is one of the most common post-translational modifications in eukaryotic cells with important roles in glycoprotein maturation and function [bib_ref] Function and 3D Structure of the N-Glycans on Glycoproteins, Nagae [/bib_ref] [bib_ref] Free oligosaccharides to monitor glycoprotein endoplasmic reticulum-associated degradation in Saccharomyces cerevisiae, Hirayama [/bib_ref] [bib_ref] C-terminus glycans with critical functional role in the maturation of secretory glycoproteins, Kawai [/bib_ref] [bib_ref] Structural insights into the glycosyltransferase activity of the Actinobacillus pleuropneumoniae HMW1C-like protein, Cioaca [/bib_ref]. Disrupted glycosylation of apoB100 may lead to its mis-folding then proteasomal degradation [bib_ref] Ectopic expression of N-acetylglucosaminyltransferase III in transgenic hepatocytes disrupts apolipoprotein B secretion..., Ihara [/bib_ref] [bib_ref] Mechanisms of glucosamine-induced suppression of the hepatic assembly and secretion of apolipoprotein..., Qiu [/bib_ref]. Therefore, the glycosylation of apoB100 may be an important mechanism for its degradation. As a secretary protein, apoB100's glycosylation may also affect its secretion. reported that the inhibition of N-linked glycosylation of apoB100 with the chemical inhibitor tunicamycin significantly inhibited apoB100 secretion in rat hepatocytes [bib_ref] The N-linked oligosaccharides at the amino terminus of human apoB are important..., Vukmirica [/bib_ref]. Furthermore, mutation in the glycosylation sites of apoB100 resulted in decreased secretion efficiency of apoB [bib_ref] Apolipoprotein B100 biogenesis: A complex array of intracellular mechanisms regulating folding, stability,..., Rutledge [/bib_ref]. It has been previously suggested that the exit of proteins from the ER is a selective process, in which transport signals present in the cytoplasmic tail of cargo membrane proteins must be recognized by coatomer proteins for their incorporation into the COPII vesicles. Two classes of ER export signals have been described for type I membrane proteins, which are diacidic and the dihydrophobic motifs. Claudio et al. proposed that the interaction of Sar1 with the [RK](×)[RK] motif at the CT of members of the glycosyl-transferase family, and probably of other type II mem-brane proteins, is an early event in the selection of these proteins as cargo of COPII-transport vesicles on their way to the Golgi complex [bib_ref] Endoplasmic reticulum export of glycosyltransferases depends on interaction of a cytoplasmic dibasic..., Giraudo [/bib_ref]. ApoB100 exits the ER in the COPII vesicles [bib_ref] Apolipoprotein B100 exit from the endoplasmic reticulum (ER) is COPII-dependent, and its..., Gusarova [/bib_ref] , thus apoB100 secretion may also be regulated by glycosylation. In this study, we found that changed expression levels of GLT8D2 protein led to similar level changes of apoB100 protein in HepG2 cells. We speculated that: (1) GLT8D2 increases apoB100 glycosylation and reduces apoB100 misfolding and proteasomal degradation;
(2) GLT8D2 increases apoB100 secretion by enhancing apoB100 glycosylation and transport from ER to trans-Golgi membrane by stimulating transport signals; (3) the effect of GLT8D2 is stronger in reducing apoB100 degradation than that in increasing apoB100 secretion. Therefore, overall, GLT8D2 positively regulates apoB100 protein expression in HepG2 cells. ApoB100 has 19 potential N-glycosylation sites, and 16 asparagine residues of ApoB have been reported to be occupied by oligosaccharides, which are high-mannose type, hybrid type, and monoantennary and biantennary complex type [bib_ref] Site-specific glycosylation analysis of human apolipoprotein B100 using LC/ESI MS/MS, Harazono [/bib_ref] [bib_ref] Apolipoprotein B, the major protein component of triglyceride-rich and low density lipoproteins, Chan [/bib_ref]. Several Ser/Thr sites may be glycosylated by O-gycans. The Glt8D2 is a glycosyltransferase gene, but its donor-substrate and receptor substrate are still unknown thus need further study.
## Experime ntal section
# Materials
## Cell culture
HepG2 cells were purchased from ATCC and reserved in our laboratory in the Institute of Infectious Disease, Capital Medical University. The Cells were cultured in Eagle's minimum essential medium (EMEM) supplemented with 10% fetal bovine serum (FBS) in a 5% CO 2 -humidified atmosphere at 37 °C .
## Total rna extraction and cdna amplification
Total RNA was extracted from HepG2 cells by using TRIzol reagent according to the manufacturer's instructions. Two micrograms of extracted RNA was reverse transcribed into cDNA using 4 μL 5× Exscript™ Buffer, 1 μL oligo dT, 1 μL Random 6mers, 1 μL Exscript™ RTase and 9.8 μL DEPC water in a final volume of 20 µL with enzyme buffer for 15 min at 37 °C . GLT8D2 cDNA was amplified by using a 25 µL reaction mixture containing 14 μL water, 5 μL 10 μM primers (2.5 μL of each), 2.5 μL buffer, 0.5 μL 10 mM dNTPs, 0.5 μL DNA polymerase, and 2.5 μL cDNA.
The specific primers used were the forward primer, 5'-GGT ACC ATG GCT CTG TTA CGA AAA ATT AAT C-3' incorporated with a KpnI restriction enzyme site, and the reverse primer, 5'-CTC GAG AGC TAT GGT GAT TGA GTT TAA ATA TC-3' incorporated with a XhoI restriction enzyme site. The GLT8D2 cDNA was amplified by using 40 cycles of denaturation at 94 °C for 40 s, annealing at 65 °C for 40 s, extension at 72 °C for 2 min, followed by 10 min at 72 °C for the final extension. PCR products were recovered from the gel, purified and sequenced.
## Plasmid construction
## Plasmid pegfp-c1-glt8d2
The purified GLT8D2 cDNA fragment was ligated into the pEGFP-C1 vector by T4 DNA ligase. Recombinant vector pEGFP-C1-GLT8D2 was transformed into competent E. coli DH5α cells. The integrity of the recovered plasmid was confirmed by KpnI and XhoI restriction enzyme digestion and sequencing.
## Plasmid pcdna6.2-glt8d2 shrna
Based on the multivariate biological information law, GLT8D2 was used as the target gene (GenBank accession No. NM.031302), and 4 pairs of short hairpin RNAs (shRNAs) and one irrelevant sequence (GLT8D2 IR) were designed [fig_ref] Table 1: shRNA Sequences. [/fig_ref]. PcDNA6.2-GLT8D2 shRNA plasmid and pcDNA6.2-GLT8D2 IR plasmid were constructed and identified by Life Technologies.
## Plasmid pegfp-c1-muta-glt8d2
To create the plasmid carrying a mutant GLT8D2 with changed glycosylation, A699 was replaced with T by site-directed mutagenesis with the primer of 5'-GTG ATT GTT GCC ATC ATG ACA GAA TGG WT-3'. The mutant GLT8D2 was amplified, purified, and ligated into pEGFP-C1 vector by T4 DNA ligase. The recombinant vector pEGFP-C1-mut-GLT8D2 was transformed into competent E. coli DH5α cells. The integrity of the recovered plasmid was confirmed by KpnI and XhoI digestion and sequencing.
## Other plasmids
The plasmids pEGFP-C1 and pDsRED-N1-calreticulin were reserved in Institute of Infectious Disease, Capital Medical University.
## Intracellular localization of glt8d2
Confocal microscopy imaging was used to detect the sub-cellular localization of GLT8D2 in HepG2 cells. HepG2 cells cultured on plates were co-transfected with pEGFP-C1-GLT8D2 plasmid (green fluorescent) and pDsRED-N1-calreticulin plasmid (red fluorescent), fixed with 4% formaldehyde 48 h post-transfection, stained with DAPI for 10 min, and examined by laser scanning confocal microscopy.
## Co-immunoprecipitation (co-ip) and western blot (wb)
For the co-IP assay, total proteins were extracted from HepG2 cells by lysis buffer, and incubated with anti-GLT8D2 and anti-apoB100 antibodies overnight at 4 °C in the presence of 50 µL Protein-G/A beads. Beads were collected, washed, and resuspended in equal volumes of 5× SDS loading buffer. The immunoprecipitated proteins were separated by SDS-PAGE and transferred onto PVDF membrane. The membrane was blocked with 5% skim milk, incubated at 4 °C overnight with anti-apoB100 antibody, and incubated with HRP-conjugated secondary antibody. The signals were analyzed using the Imaging System.
## Detection of apob100 expression in hepg2 cells
2 × 10 5 HepG2 cells were seeded onto 6-well plates and grown overnight at 37 °C. The plasmids pEGFP-C1-GLT8D2, pcDNA6.2-GLT8D2 shRNA-1 (result suggest that pcDNA6.2-GLT8D2 shRNA-1 has the best inhibition effect on GLT8D2 mRNA expression in HepG2.), pEGFP-C1-muta-GLT8D2 and pEGFP-C1 empty vector was transfected into HepG2 cells by using Lipofectamine 2000 reagent according to the manufacturer's instructions. After 48 h, proteins were extracted from HepG2 cells by lysis buffer and analyzed with Western-blot.
## Statistics
Data are expressed as mean ± SEM. Statistical significance was assessed by two-way ANOVA. A difference with p value less than 0.05 was considered as statistically significant.
# Conclusions
In this study, we cloned GLT8D2 gene from HepG2 cells and generated a series of plasmids for in vitro studies of its molecular functions. We discovered that (1) GLT8D2 is localized in the ER;
(2) GLT8D2 has interaction with apoB100; and (3) GLT8D2 positively regulates the levels of apoB100 protein in HepG2 cells. Based on these results, we propose that GLT8D2 is a glycosyltransferase of apoB100 that regulates apoB100 levels in hepatocytes.
[fig] Figure 1: Electrophoresis of GLT8D2 PCR product on 1% agarose gel. Lane 1: DNA marker; Lane 2: Negative control; Lane 3: amplified GLT8D2 gene (1050 bp). [/fig]
[fig] Figure 2: Electrophoresis of recombinant GLT8D2 on 1% agarose gel. Lane 1: DNA marker; Lane 2: Double digestion of recombinant pEGFP-C1-GLT8D2 with KpnI and XhoI restriction enzymes (pEGFP-C1, 4700 bp and GLT8D2, 1050 bp). [/fig]
[fig] Figure 3: Knockdown efficiency of GLT8D2 shRNAs in HepG2 cells revealed by qRT-PCR. [/fig]
[fig] Figure 4: Co-localization of GLT8D2 and calreticulin in HepG2 cells. (A) Calreticulin with red fluorescence; (B) GLT8D2 with green fluorescence; (C) Nuclear with blue fluorescence; (D) Calreticulin and GLT8D2 merge with yellow fluorescence. [/fig]
[fig] Figure 6: Western blot analysis of apoB100 expression in HepG2 cells with changed expression of GLT8D2. Lane 1: pEGFP-C1-GLT8D2; Lane 2: pcDNA6.2-GLT8D2-shRNA-1; Lane 3: pEGFP-C1-muta GLT8D2; Lane 4: pEGFP-C1 empty vector; Lane 5: pcDNA6.2-GLT8D2 IR; Lane 6: Control. [/fig]
[table] Table 1: shRNA Sequences. [/table]
|
Cancer-Associated Fibroblasts and Tumor-Associated Macrophages in Cancer and Cancer Immunotherapy
Our understanding of the tumor microenvironment (TME), including the interplay between tumor cells, stromal cells, immune cells, and extracellular matrix components, is mandatory for the innovation of new therapeutic approaches in cancer. The cell-cell communication within the TME plays a pivotal role in the evolution and progression of cancer. Cancer-associated fibroblasts (CAF) and tumor-associated macrophages (TAM) are major cell populations in the stroma of all solid tumors and often exert protumorigenic functions; however, the origin and precise functions of CAF and TAM are still incompletely understood. CAF and TAM hold significant potential as therapeutic targets to improve outcomes in oncology when combined with existing therapies. The regulation of CAF/ TAM communication and/or their differentiation could be of high impact for improving the future targeted treatment strategies. Nevertheless, there is much scope for research and innovation in this field with regards to the development of novel drugs. In this review, we elaborate on the current knowledge on CAF and TAM in cancer and cancer immunotherapy. Additionally, by focusing on their heterogenous functions in different stages and types of cancer, we explore their role as potential therapeutic targets and highlight certain aspects of their functions that need further research.
# Introduction
Originating from the neighboring healthy tissues and recruited from the circulation, a multitude of proliferating non-neoplastic cells such as fibroblasts, macrophages, immune cells, and endothelial cells contribute to carcinogenesis within the tumor microenvironment (TME). Cancerassociated fibroblasts (CAF) and tumor-associated macrophages (TAM) are the major cell populations within the stroma of all solid tumors in which they often exert protumorigenic functions. Although their precise interactions remain to be elucidated, CAF and TAM strongly modulate disease progression, therapy resistance, and clinical outcomesand may function in synergy.
Targeting the cytokines, inhibitory immune checkpoint ligands expressed by CAF and TAM, and antiphagocytic signaling by tumor cells have shown some efficacy in preclinical trials. The results of clinical trials are nonetheless ambiguous. Antibodies, chemokines, and chemokine ligands that interfere with CAF/TAM interactions, and their combinations hereof, are highly prioritized in experimental clinical regimens that are aimed at modulating the TME.
## The fibroblast
Fibroblasts can be clearly identified and characterized by their elongated morphology, the lack of epithelial, endothelial, leukocytic, and malignant-cell markers, and the positivity for mesenchymal markers such as vimentin. Under normal circumstances, fibroblasts are present in abundance in the connective tissues in a dormant state, transiently being activated during periods of tissue remodeling and repair. They are involved in the production of extracellular matrix (ECM) and modulation of inflammation, as well as the proliferation and differentiation of epithelial cells.
Well-established fibroblast-activating signals include inflammatory mediators, transforming growth factor-beta (TGF-b), and lysophosphatidic acid which increase the activity of SMAD transcription factors and drive the expression of alphasmooth muscle actin (a-SMA) that provides the fibroblast with a highly contractile phenotype (usually known as myofibroblast or a-SMA + fibroblast). The activated fibroblasts produce chemokines and cytokines to regulate the communication with other mesenchymal, epithelial, and immune cells. Importantly, all of these functions are utilized and enhanced in cancer.
## Cancer-associated fibroblasts
## Structure and functions
Within a tumor, the mesenchymal cells that comply with the aforementioned definitions above, are generally referred to as CAF. Compared with regular fibroblasts, they tend to be slightly larger with darker nuclei and branched cytoplasm. CAF may differentiate from quiescent fibroblasts and bone marrowderived mesenchymal stem cells or trans-differentiate from epithelial cells, smooth muscle cells, pericytes, and adipocytes.
CAF are present during all stages of solid malignanciesand their functional impact on the biology of cancer is assumed to be similar across all tumor types.
CAF are the predominant cell type in the tumor stroma and they contribute to the proliferative, pro-inflammatory, immunosuppressive, angiogenic, pro-invasive, and pro-metastatic TME that is required for the evolution and progression of cancer.
Inflammatory mediators such as TGF-b, interleukin (IL)-1, and IL-6 produced by tumor cells and non-malignant stromal cells promote CAF activation and contribute to a proinflammatory profile, that directly support carcinogenesis. The activation of specific transcriptional programs and the lack of negative feedback mechanisms launch CAF into selfsustaining trajectories.
Additionally, CAF drive the epithelial-mesenchymal transition (EMT), whereby cancer cells lose polarity and adhesion molecules and gain the motility necessary for dissemination. Despite the overall pro-tumorigenic effects, functional dualities have been observed. A hypothesis is that initially CAF are tumor suppressive but as cancer evolves they transform into pro-tumorigenic cells.
## Heterogeneity of caf subtypes
Within a multi-clonal solid tumor, CAF are differentially exposed to a multitude of tumor secreted factors (TSF) explaining their heterogeneity. However, the essential molecular mechanisms underlying the activation and pro-tumorigenic activities of fibroblasts may be common to various cancers, which present a manifold of targets for innovative CAF-targeted therapies. Signaling cascades mainly involve the Wnt/b-catenin, TGF-b, epidermal growth factor receptor, JAK/STAT, and Hippo pathways.
Several studies have characterized distinct CAF subgroups that differentially express the CAF markers, e.g. a-SMA, fibroblast activation protein (FAP), and platelet-derived growth factor receptor (PDGFR), and show that CAF subpopulations may have various and even opposing functions. Tumor-suppressive CAF populations have been characterized by activated Hedgehog signaling pathways in mouse models of colon, pancreatic, and bladder cancers. However, the full complement of CAF populations remains unclear, and more detailed classifications and functions of CAF subtypes are needed.
In a mouse model of pancreatic ductal adenocarcinoma (PDAC), the ablation of CAF led to enhanced hypoxia, EMT, increased vascularity, cancer cell proliferation, and disease progression demonstrating that CAF to some extent can restrain tumor growth. Similarly, an initial expansion of local fibroblasts circumscribing early or premalignant lesions in response to tissue neoplasia was observed in mouse models and human tissue studies.
Thus, the TME comprises a heterogeneous population of CAF subtypes or clusters with different functions associated with immunomodulation, immunosuppression, and immunotherapy resistance.
Furthermore, in a mouse model on early and late PDAC stages, fibrosis associated with type I collagen provided a protective response from the host rather than a pro-tumorigenic response.
These results demonstrate that at least some stromal constituents may restrain rather than promote tumor progression and illustrate the high degree of temporal differentiation plasticity within the diverse cell populations of tumors. This may also explain the conflicting reports regarding antitumor and pro-tumor functions of CAF.
In a preclinical trial on lung cancer, the depletion of CAF significantly reduced the number of metastases. To establish the clinical relevance of primary tumor CAF in the formation of metastasis, this research group examined human brain metastases (since the normal brain is devoid of fibroblasts) from lung, breast, kidney, and endometrium, and found a distribution of activated CAF within these metastases. These findings support the view that the CAF shed from the primary tumor, together with cancer and non-tumor cells from the TME, survive during the blood circulation and proliferate at the metastatic site.
With respect to human PDAC specimens, the patients with a higher expression of FAP were found to be associated with shorter disease-free survival and overall survival when compared to those with low FAP expression. The immune suppression caused by FAP + CAF is mediated by the CXCL12 receptor CXCR4 that excludes T cells from the tumor. Notably, CXCR4 inhibition leads to an elimination of tumor cells by a rapid accumulation of cytotoxic CD8 + T cells. Moreover, the deletion of FAP + CAF using a FAP-targeted immune-based therapeutic approach or a genetic ablation approach inhibited cancer growth in murine PDAC models. Thus, the inhibition of CAF-induced protumorigenic signals is a highly attractive future strategy to improve outcomes in pancreatic cancer.
In human triple-negative breast cancer, a subset of CAF with myofibroblast characteristics (myCAF) (a-SMA + /FAP + or S1 CAF) was identified as a key player in immunosuppression through the attraction of T regs and inhibition of effector T cell proliferationand it was hypothesized that targeting the CAF-S1-mediated immunosuppression could enhance antitumor immunity.
In PDAC, CAF are linked to worse overall survival. PDAC is infamous for the abundance of fibrotic ECM with the majority of the tumor volume being composed of a-SMA + CAF. Preclinical and clinical trials targeting stromal a-SMA + CAF, however, resulted in an apparent, paradoxical acceleration in disease progression and reduction in survival, halting clinical trials and adding further layers of complexity to CAF functions.
Another study on murine models of lung carcinoma and PDAC revealed that the deletion of FAP led to a significant reduction in CAF infiltration and tumor tissue necrosis, and an increase in infiltration of CD8 + T cells. Moreover, in murine models of breast and colon cancer, the administration of a DNAbased vaccine targeting FAP induced the killing of CAF by CD8 + T cells and lead to a substantial increase in the uptake of chemotherapeutic agents by otherwise multi-drug-resistant cancer cells. Further, in immunocompetent mice, the cell transfer of FAP-specific chimeric antigen receptor T cells boosted host immunity and arrested pancreatic tumor growth; however, it also led to significant lethal toxicity and cachexia. These examples indicate that specific CAF subsets could be potential targets for improving immunotherapy. Future studies are needed to develop targeted therapies aimed at specific CAF populations.
## Secreted factors and exosomes in caf-tumor cells interplay
The cytokines and chemokines produced by CAF may have both immunosuppressive and immuno-activating effects on various leukocytes, including CD8 + T cells, immunosuppressive regulatory T cells (T regs ), and macrophages. However, the consensus is that the overall effects of CAF are immunosuppressive. IL-6, CXC-chemokine ligand (CXCL) 9, and TGF-b, which are produced by CAF, have well-established roles in suppressing anti-tumor T cell responses. This is also supported by an inverse association between CAF and CD8 + T cell cytotoxicity.
The staining of the inhibitory immune-receptor ligand programmed death-ligand 2 (PD-L2) and tumor necrosis factor-alpha (TNF-a) ligand OX40L in human breast cancer sections revealed T lymphocytes at the surface of CAF. This confirmed that subsets of CAF attract and retain T lymphocytes at the periphery of the tumor through distinct mechanisms involving chemokine signaling (chemokine ligand [CCL]-11, CXCL12-14), cell adhesion molecules, activation of inhibitory immune checkpoints, and CD8 + T cell anergy.
In a murine PDAC model, it was demonstrated that CAF, programmed by TGF-b to express a leucine-rich protein (LRRC15), were associated with a poor response to anti-PD-L1 therapy. Additionally, CAF are a source of various growth factors including TGF-b, vascular endothelial growth factor (VEGF), fibroblast growth factor 5, growth differentiation factor 15, hepatocyte growth factor and insulin-like growth factor. The secretion of pro-stemness paracrine factors such as insulin-like growth factors, inflammatory cytokines (IL-6 and IL-8), and chemokines (CCL2 and CCL5) promotes the conversion of cancer cells into cancer stem cells and reinforce the stemness of existing cancer stem cells. Moreover, the secretion of IL-6 make CAF an important mediator of EMT in cancer cells.
Exosomes are extracellular vesicles released by all cell types and are found in all bodily fluids (50). They contain genetic material, proteins, and lipids and are essential for intercellular communication. The activation, recruitment, and conversion of fibroblasts into activated CAF depend on TSF and tumorsecreted exosomes (TSE) containing various oncogenic molecules such as microRNAs (miRs), fusion gene mRNAs, long non-coding RNAs, mutated DNA fragments, and a manifold of cell-signaling molecules. The circulating levels of exosomal miRNA accurately reflect disease progression and could serve as a prognostic tool among various cancers following resection of the primary tumor.
In addition to TSF, TSE and CAF-derived exosomes (CAFEx) secreted by tumor cells and CAF, respectively, in the primary tumor are critical mediators of cancer cell-immune cell communication and they drive the formation of pre-metastatic niches (PMN). Moreover, CAF may enter the circulation and promote the development of PMN and subsequent metastatic lesions.
Integrins (ITG) are known to determine tumor cell organotropism. In a mouse model, CAF promoted lung metastasis by the construction of PMN via CAFEx. CAFExderived ITGa 2 b 1 were found to home to the lung fibroblasts and subsequently activate the TGF-b signaling pathway. To prepare for subsequent colonization of the lung tissue by extravasating circulating tumor cells (CTC), the lung microenvironment is remodeled by the activated lung fibroblasts. Surface ITG guide the TSE to organ-specific ECM ligands (collagen, fibronectin, fibrinogen, and E-cadherin) in the target organs, e.g. ITGa 6 b 1 and ITGa 6 b 4 adhere to the epithelial cells and fibroblasts in the lung and ITGa v b 5 binds to resident liver macrophages (Kupffer cells) and upregulate the genes for cell migration and S100 protein. Organ-specific TSE have been identified for 28 different metastatic cell lines. Furthermore, TSE comprising TGF-b and PDGF mediate the activation, differentiation, and recruitment of CAF through all stages of all solid cancers.
In early-stage colorectal cancer (CRC), TSE were found to promote highly proliferative and angiogenic CAF, while those from late-stage metastatic CRC cell lines were observed to induce highly invasive CAF which, through the secretion of ECMdegrading proteases and increased expression of the proinvasive modulators of membrane protrusion, enabled the penetration of ECM.
In addition, TSE alter CAF metabolism and induce the production of CAFEx containing nutrient metabolites (amino acids and tricarboxylic acid cycle intermediates) that fuel the tumor cells and increase their survival. A study on breast-cancer cell lines revealed that TSE containing miR-105 could re-program CAF metabolism and enable them to increase glucose metabolism when nutrient levels were sufficient as well as detoxify metabolic wastes into energy-rich metabolites when nutrients were scarce.
As shown in PDAC, lactate produced by cancer cells promotes extensive epigenomic reprogramming of CAF. In CRC, and during protein deprivation, CAF accumulate fatty acids, phospholipids, and fatty acid synthetase. The uptake of lipid metabolites by the CRC cells secreted by CAF seem to be essential for their migration.
Another potent promotor of malignancy is the heat shock factor 1 which is frequently activated in CAF. It drives a program that supports the survival and metastatic potential of cancer cells by inhibiting apoptosis and promoting migration. The activation of heat shock factor 1 has been associated with poor outcomes in CRC, lung-, breast-, and hepatocellular carcinoma (HCC).
Of the important players, the gene that deserves mentioning is the HMG-box 2 (SOX2). It codes for transcription factors controlling the expression of several genes involved in early embryonic development. The upregulated stromal SOX2 drives the reprogramming of colonic fibroblasts that results in enhanced b-Catenin and TGF-b signaling in CRC cells supporting cancer progression. Nonetheless, the precise mechanism remains to be determined.
The subset of CAF with myofibroblasts characteristics (myCAF) mediate a chronically deranged wound healing program in tumors and play a key role in the development of a continuously evolving fibrotic stroma. myCAF are highly responsive to chemokines and metabolically and morphologically distinctive from CAF. When activated, their proliferation rate drops and the production of ECM components increases dramatically. The cytoplasmic microfilaments of myCAF connect to the extracellular fibronectin domains, creating very contractile mechanisms. The following extracellular deposition of collagen reinforces and stiffens the ECM. Not only does it contribute to the increasing stromal density, but the remodeling of the stroma by CAF-produced matrixenzymes also provides tracks for cancer cell invasion and migration. The stromal stiffness results in increased interstitial pressure, abnormal vasculature, collapsed blood vessels, hypoxia, and acidity which lead to inefficient drug delivery and reduced response to therapy. These physical and chemical barriers are hostile to cytotoxic immune cells such as CD8 + T cells and natural killer (NK) cells.
## Caf and circulating tumor cells (ctc)
The presence of CAF in the circulation of cancer patients and their levels in the peripheral blood correlates with cancer progression and worse prognosis. Notably, the high levels of CAF-CTC aggregates in the blood samples from patients should be considered an important marker of worse clinical outcomes. For instance, CTC have higher viability in the blood stream when accompanied by stroma cells that also provide an advantage with respect to early survival and growth of tumor cells at the metastatic site. Traveling in clusters with macrophages, immune cells, and platelets, CAF support, shield, and increase the survival of CTC. Adjoining neutrophils may aid in the survival of CTC through the suppression of leukocyte activation. Through strong intercellular adhesions, CAF maintained the viability and proliferative capacity of CTC in cellular aggregates in presence of high levels of hemodynamic forces (> 1,000 dyn/cm 2 ). This protective role was observed in prostate cancer, usually spreading through blood vessels rather than the lymphatic system.
Only a minority of CTC travel in clusters; however, in a mouse model, it was estimated that the probability of metastasis formation originating from clusters (and especially those of oligoclonal tumor cell groupings) is fifty times higher compared with that originating from a single CTC.
As EMT of tumor cells may proceed within the clusters, the association between neutrophils and CTC drives tumor cell mitosis and expands the metastatic potential of CTC. Upon arrival in the PMN, tissue-resident fibroblasts contribute to the mesenchymal-epithelial transition (MET). Thus, CAF are considered key players in promoting the survival of CTC.
## Targeting caf-associated pathways
To revert CAF to a quiescent state by targeting the activation pathways is an appealing concept. CAF-secreted Wnt2 accelerates the Wnt/b-catenin signaling pathway which corresponds with the absence of CD8 + T cells. The effects of vitamin D seen in epidemiological studies of PDAC and CRC are partly related to the reduced CAF-related Wnt/b-catenin signaling which was relayed by vitamin D metabolites.
Alternatively, targeting CAF-derived cytokines and chemokines (e.g. CXCL, IL-6, and TGF-b) could improve anticancer efficiency in combination with immunotherapy. Several IL-6 inhibitors such as sarilumab and tocilizumab that are already approved for autoimmune and myeloproliferative disorders, are being investigated for their role in anticancer therapy either alone or in combination.
Anti-TGF-b in combination with anti-PD-L1 antibodies inhibited TGF-b signaling in CAF and facilitated T cell penetration into solid tumors. A summary of RCT examining the effects of targeting IL-6 and TGF-b have been presented in. The complexity and incomplete understanding of CAF functions necessitate further research before anti-CAF targeted therapy can be integrated into clinical practice.
## The macrophage and its m1 and m2 subtypes
Representing another major stromal cell population, macrophages are remarkable, heterogenic, and versatile cells. These cells are capable of switching functions and phenotypes, depending on their unique microenvironment. They engulf; and contribute to tissue remodeling, angiogenesis, and homeostasis. The conventional binary model distinguishes between the M1 and M2 macrophages. The M1 subtype consists of classically activated, proinflammatory macrophages with bactericidal, tumor-suppressive, and anti-angiogenic functions. They express inducible nitric oxide synthase (CD86 and CD169) and are activated through their pattern recognition receptors upon recognition of damage-or pathogen-associated molecular patterns such as bacterial lipopolysaccharides and DNA damage. They produce inflammatory cytokines (e.g. IL-1b, IL-6, IL-12, IL-23, and TNF-a), proliferate, and self-renew in a macrophage colonystimulating factor 1 (M-CSF1)-and granulocyte-macrophage (GM)-CSF-dependent manner.
The M2 subtype, the alternatively activated macrophages expressing CD163, CD206, and CD204, are commonly known as TAM. They are characterized by the production of antiinflammatory, immunosuppressive chemokines and cytokines, such as IL-4, IL-6, IL-8, IL-10, IL-13, and TGF-b, and are devoid of cytotoxic activity. They produce various growth factors, such as basic fibroblast growth factor (b-FGF), placental growth factor, insulin-like growth factor, epidermal growth factor (EGF), VEGF, and PDGF.
It should be emphasized that macrophages are extremely plastic. Many context-and tissue-dependant phenotypes on the spectrum between M1 and M2 exist, depending on multiple factors of stimulation, and these in-between phenotypes are not captured by the classical nomenclature. A more comprehensive classification system that takes the dynamic nature of macrophages into account has been proposed but so far not adopted in the literature.
Although their origin is still debated, it is generally believed that macrophages originate via common dendritic cell precursors in blood, spleen, and from bone marrow hematopoietic stem cellderived progenitors with myeloid restricted differentiation. Embryonic precursors may seed tissues already in the fetal period and become tissue-resident macrophages. Attracted by chemokines, macrophage progenitors enter the circulation from reservoirs in the bone marrow and spleen. They leave the peripheral blood flow and migrate to tissues where local growth factors and cytokines control their differentiation.
## Tumor-associated macrophages
The level of infiltrating TAM correlates with tumor progression and reduced survival in patients. Growth factors and immunosuppressive cytokines produced by TAM can enhance motility, intravasation, and invasion of tumor cells, as well as stimulate angiogenesis and prevent attacks by T cells and NK cellsas observed in various tumor types including carcinomas, sarcomas, and lymphomas. The recruitment of macrophages and their differentiation into TAM are primarily promoted by TSF and CAF-derived factors such as M-CSF1, GM-CSF, CCL2, VEGF, IL-6, and IL-8; and are related to local anoxia, acidity, and inflammation. The infiltration into the TME is determined by CC chemokines such as the C-C motif ligands CCL2, CCL11, CCL16, and CCL21 produced by local lymphatic endothelial cells and stromal cells as demonstrated in breast-, lung-, oesophageal-, ovarian-, and cervical cancers. Especially CCL2 exhibits strong chemotactic activity for macrophages. Producing CCL2 themselves, macrophages recruit macrophages in a feed-forward loop.
Homing towards increasing gradients of chemotactic molecules, TAM massively infiltrate hypoxic/necrotic regions of tumors and survive by shifting their metabolism towards glycolysis. Hypoxic TAM express the transcription factor hypoxia-inducible factor 1a and secrete VEGF, b-FGF, PDGF, cyclooxygenase-2, prostaglandin E2, and MMPs. In response to hypoxia, TAM also overexpress PD-L1, PD-L2, and cytotoxic T-lymphocyte-associated protein 4 ligands that contribute to immune cell dysfunction and limit the effects of checkpoint inhibitors. Furthermore, the high levels of IL-10 and TGF-b produced by TAM block T cell proliferation and T cell cytotoxicity, while activating T regs.
## Exploring tam in different cancer types
Activated TAM are significant prognostic biomarkers for breast cancer, PDAC (105), non-small-cell lung cancer (106), gastric cancer (107), HCC (108), and stage II colon cancer.
In breast cancer, TAM produce metalloproteinases (MMP) and cathepsins which degrade the ECM and release angiogenic factors stored in the ECM. TAM-derived MMP-2 and MMP-9 have been correlated to a worse prognosis. Using human metastatic breast cancer cells, it was demonstrated that these cells stimulate TAM with M-CSF1 and in turn, TAM supply EGF to them. This paracrine feed forward mechanism between tumor cells and TAM facilitates the dissemination, intravasation, and metastatic spread of cancer. In gastric cancer, TAM-derived exosomes that are rich in miRNA, lncRNA, and specific proteins contribute to tumor cell dissemination. Mass spectrometric analysis revealed that these exosomes activated mitogenic signaling through the phosphatidylinositol 3-kinase (PI3K)/AKT pathway in tumor cells, inducing EMT and increasing the metastatic potential.
In PDAC, TAM-derived exosomes reportedly contribute to the resistance of tumor cells to gemcitabine. Using a genetic mouse model of PDAC and electron microscopy analyses, it was demonstrated that TAM exosomes are selectively internalized by tumor cells indicating that TAM and tumor cells communicate closely with each other. Furthermore, it was shown that the sensitivity of PDAC to gemcitabine was significantly reduced by the exosomal TAM-derived miR-365.
In non-small-cell lung cancer tissue samples from 104 patients, M1 macrophages and TAM were identified using multiplex immunofluorescence staining. TAM predominated over M1 macrophages in number and proximity to tumor cells, which was linked with tumor cell survival, particularly in the hypoxic regions.
In stage II colon cancer, postoperative adjuvant chemotherapy generally has limited effect, with an improved survival rate of less than 5% at 5 years after surgery. In a clinical study on human stage II colon cancer, a high density of CD206 + TAM was significantly associated with poor differentiation and worse disease-free survival. A high CD206/ CD68 ratio (CD68 being an unspecific marker for the macrophage lineages) was significantly associated with poor differentiation, T4 stage, and lymphatic/vascular/perineural invasion. This ratio was a more reliable prognostic factor than CD206 + TAM density and other traditional clinicopathologic high-risk factors. Notably, the CD206/CD68 ratio identified patients with a low and high risk of tumor recurrence and effectively predicted which patients would benefit from adjuvant chemotherapy.
## Targeting tam-associated pathways
The field of research exploring the mechanisms by which TAM impact the tumor progression and lower the response to anticancer therapies is very active, and it includes several pharmacological strategies to target TAM. While some strategies revolve around blocking recruitment and depleting TAM through direct inhibition using small molecules and monoclonal antibodies, others focus on reprogramming of TAM. With respect to TAM recruitment, it has been demonstrated that the blockade of CCR2 suppresses the accumulation of TAM in tumors. CCR2 inhibitors and anti-CCL2 antibodies (CNTO 888) have demonstrated efficacy in reducing tumor growth and metastasis in several pre-clinical murine models.
It has been reported across multiple murine tumor-and metastasis models that CCR2 antagonism in combination with anti-PD-1 therapy lead to sensitization and enhanced tumor response over anti-PD-1 monotherapy. Additionally, in a clinical trial on PDAC, an objective tumor response was observed in 16 of the 33 patients (49%) receiving a CCR2 inhibitor (PF-04136309) plus FOLFIRINOX, compared to FOLFIRINOX alone(121). CCR5 is another receptor which is highly upregulated in metastatic cancers, and a study in mice showed promising response upon treatment with CCR5 antagonist, maraviroc.
Furthermore, several trials investigating the effect of dual inhibition of CCR2 and CCR5 in patients with locally advanced pancreatic cancer, CRC, HCC, advanced renal cell carcinoma and non-small-cell lung cancer are underway.
Another strategy is to deplete TAM by pharmacological blockade of CSF and its receptor CSF-1R, in mono-or combination therapy, preferentially in patients with advanced solid tumors. The depletion of TAM by CSF-1R blockade showed increased infiltration of CD8+ cytotoxic T cells and improved treatment response in murine models of breast, prostate, and cervical tumors. Inhibition of the CSF-1/CSF-1R axis, using antibodies (AMG 820, IMC-CS4) and small molecule inhibitors like pexidartinib, is presently being explored in phase I/II clinical trials.
Additionally, the plasticity of macrophages opens up new avenues for reprogramming TAM to switch to an anti-tumor, M1-subtype. While drugs targeting toll-like receptors (imiquimod) are already approved for use, many novel antibodies and fusion proteins targeting CD47/SIRPa axis are under investigation. Adding to that list, some preclinical trials are currently investigating the use of CAR-T adoptive cell transfer and mRNA tumor vaccines. Theoretically, strategies to reprogram TAM by the delivery of mRNA are attractive, but this research is still in its nascent stages.
To this end, TAM are a promising therapeutic target and further research will benefit in the development of combinational regimens utilizing multifaceted targeting of the cancers.
## The caf-tam collaboration
Although CAF and TAM can play both supportive and restrictive roles in carcinogenesis and tumor progression, they are emerging as key players in orchestrating cancer-promoting inflammation and their interactions likely increase the malignancy of tumors.
Further to the recruitment of monocytes and M2 polarization, recent data have linked CAF and TAM to a reciprocal interplay with cancer cells. The anti-inflammatory and immunosuppressive M2 phenotype facilitates tumor growth and converts healthy fibroblasts into CAF. Activated CAF secrete factors that promote TAM, cancer cell aggressiveness, EMT, and stemness. In return, cancer cell secrete factors that increase CAF activation and reactivity in a complex that involves various interleukins, chemokines, growth factors and proteinases.
As the synergistic interaction between TAM and CAF was only recently identified, only a few studies describe their cell-cell interactions. In CRC and oral squamous cell carcinoma, high levels and combined presence of CAF and TAM within the TME was reported as a negative prognostic factor. In high-risk neuroblastoma, pro-inflammatory lipid mediators produced by CAF contributed to tumor growth and were accompanied by a high infiltration of CD163 + TAM.
The CAF secretome seems to regulate the composition of tumor-related inflammation, including the presence, phenotypes, and levels of infiltrating TAM. In this case, CAF together with tumor cells shape the environment to which monocytes/ macrophages are recruited to promote tumor progression. To evaluate the effects of CAF on tumor growth and metastasis, monocytes were co-cultured with colon cancer cells and stimulated with colon cancer-activated CAF. The inducible factors that drove monocyte differentiation into pro-invasive TAM were primarily characterized as CAF-derived GM-CSF and IL-6, and are known to regulate the presence of TAM and promote cancer cell invasion and metastasis. Therefore, in the triple crosstalk between tumor cells, CAF, and TAM, IL-6 and GM-CSF could become important targets for modulating their interaction. In HCC, osteopontin (OPN) was identified as a key molecule involved in cancer-CAF-TAM interactions. OPN is a chemokine-like phosphorylated glycoprotein released by TAM in the TME. The TAM-secreted OPN promotes the secretion of OPN from CAF and leads to increased cancer cell malignancy through upregulation of proliferation, ECM degradation, and migration. Thus, OPN could be a potential new therapeutic target to inhibit cancer-CAF-TAM interactions in HCC.
Based on global gene expression profiles in CRC, bioinformatics and immunohistochemistry identified stromal markers that were significantly associated with resistance to therapy, recurrence and poor prognosis. The predictive power of stromal cell genes was higher than the power of tumor cell genes. In accordance with and by investigating the four consensus molecular subtypes (CMS) in CRC, the CMS4 tumors were characterized by heavy infiltration of mesenchymal cells and displayed worse recurrence-free survival and overall survival compared with other CMS subtypes. Moreover, the CMS4 tumors showed a clear upregulation of genes controlling EMT, TGF-b signaling, angiogenesis, matrix remodelling, and inflammation. Compared with TAM, and playing a dominant role in the evolution of TME, a higher density of CAF is usually observed in tumors of the gastrointestinal tract, pancreas, lung, and prostate. It is important to mention that TAM are associated with migration and intravasation of tumor cells, CTC formation, and aiding CTC clusters in the peripheral circulation in the patients. Despite them being appealing targets, owing to the lack of selectivity, strategies to attack CAF and TAM have resulted in unwanted sideeffects and thereby, limited their clinical use.
## The extracellular matrix
The ECM mainly consists of proteins and glycosaminoglycans that are constantly remodeled by fibroblasts and macrophages in response to environmental changes.
In preclinical trials on breast and lung cancer, it was reported that CAF-produced collagen and CAF-derived FAP transformed the ECM into an environment facilitating the cancer cell motility through a parallel alignment of the collagen fibers that enhanced the direction and speed of the migrating cells.
Regular tissue fibroblasts synthesize and release ECM components such as collagen, elastin, fibronectin, and a variety of proteoglycans that combine to form a web of fibers. This network regulates the homeostasis of cells, tissues, and organs and allows the ECM and tumor cells to resist a wide range of chemical and mechanical stress factors.
In a solid tumor, the assembly of ECM fibrils is crucial for the barrier formation and exclusion of immune cells and therapeutics. Further, the collagen network in the stroma is key for the maintenance and exchange of fluids and solutes within the tumor.
Elastin, an abundantly expressed protein in the ECM, is secreted by fibroblasts as a precursor protein, tropoelastin, which assembles in the elastic fibers that are rich in crosslinks. The crosslinks render the elastin insoluble and equip the fibers with the ability to withstand repeated distension. Additionally, the elastin fibers are tightly associated with collagen fibrils which are mediated by the cell surface proteoglycans.
Fibronectin, also secreted by fibroblasts, binds to the ECM components such as collagen and fibrin and anchors the fibrils to the cell-surface integrin receptors.
The tyrosine kinase inhibitor, Imatinib-specific to ABL1, PDGFR, and c-kit-is used to treat hematological malignancies and gastrointestinal stromal tumors. It is found to increase the flow of fluids through the interstitial compartment of the tumor, improving drug delivery, mainly due to a decreased collagen fibril diameter. CAF and TAM produce various enzymes, including matrix metalloproteinases (MMP), fibrinolysin, and cathepsins that degrade ECM components, accelerate local invasion of tumor cells, and facilitate their dissemination. Some ECM degradation fragments may even stimulate angiogenesis and migration. MMPs are zinc-dependent ECM-remodelling endopeptidases deeply implicated in almost all steps of metastasis. A high MMP expression in the tumor correlates with poor prognosis and increased risk of recurrence. The CAF expression of MMP-11 in CRC, MMP-2 and MMP-9 in breast cancer, and MMP-21 in HCC was significantly related to a high risk of tumor recurrence.
The presence of hypoxia, acidity, increased interstitial pressure, and aberrant vasculature in the TME confer tumor cells with a survival advantage. The environment inhibits the penetration, navigation, and functionality of cytotoxic immune cells in their quest to kill tumor cells.
To prevent intracellular acidity, tumor cells express various proton flux regulators, such as H + -ATPases, Na + /H + exchangers, monocarboxylate transporters, carbonic anhydrases, and Na + / HCO 3 transporters. Proton pump inhibitors are currently being used in clinical trials, in combination with therapies targeting carbonic anhydrases: Acetazolamide (carbonic anhydrase inhibitor) and radiotherapy for small cell lung cancer (NCT03467360), carbonic anhydrase IX inhibitor and Gemcitabine (antimetabolite) for PDAC (NCT03450018), and Acetazolamide and Temozolomide (alkylating agent) for malignant glioma of the brain (NCT03011671). Additional clinical trials of therapies that aim to target ECM and ECMassociated molecules are on-going; however, as therapeutics, ECM degrading agents must be used with caution as they may have fundamental consequences on cell and tissue functions, which could ease the metastatic spread instead of inhibiting tumor progression.
## Caf and tam in immunotherapy and anti-angiogenesis
The introduction of monoclonal antibodies targeting inhibitory receptors on immune cells, known as immune checkpoint inhibitors, has been a great breakthrough in oncology, immensely improving the clinical outcomes of several cancers. This therapeutic strategy enhances the efficacy of anti-tumor immune responses and revitalizes exhausted killer cells such as CD8 + T cells and NK cells.
The exclusion of immune cells from solid tumors is not only caused by the physical and chemical barrier of the ECM, but also by the immune checkpoint ligands expressed by cancer cells, CAF, and TAM. In line with this, a study on tissue samples from patients with PDAC demonstrated that PD-L1 and PD-L2 (both ligands to PD-1) expressed by CAF were involved in immune cell exclusion and anergy.
Adding to the complexity of stromal cell functions, preclinical studies suggest that some CAF, along with normal fibroblasts, have the ability to overrule oncogenic signaling from the surroundings and act as tumor suppressors. Whether these fibroblasts are subtypes of normal fibroblasts resistant to CAF conversion or distinct anti-tumor CAF subpopulations remains unknown. However, the CAF/TAM collaboration do play a vital tumorpromoting role. It fuels the growth of tumors; induces stemness and EMT in cancer cells by the production of cytokines, chemokines, e.g. interleukins, TGF-b, CCL, and CXCL chemokines. It supplies the tumors with energy-rich metabolites and upregulate the tumor-cell mitochondrial oxidative phosphorylation. Thus, therapeutic regimens targeting the TAM-CAF interaction in combination with immunotherapy could improve anti-tumor therapeutic efficacy.
CSF-1R receptors are overexpressed on TAM in many cancers, controlling the production, differentiation, and function of macrophages. In a mouse model, a CSF-1R-inhibitor blocked the production of inflammatory mediators in TAM, inhibited the recruitment of bone marrow-derived suppressor cells (BMDSC), and enhanced T-cell infiltration and CD8 + T cell activity. However, the inhibition of CSF-1R signaling caused CAF to secrete chemokines and chemokine ligands that neutralized the CSF-1R inhibitor. The supplementation of a chemokine receptor antagonist reduced the tumor burden, and tumor growth was completely blocked when an immune checkpoint inhibitor (anti-PD-1) was further added to the combination.
There are currently several clinical trials evaluating the effect of CSF1R monoclonal antibodies in combination with immune checkpoint inhibitors in a variety of solid tumors.
In a human trial on solid tumors, dual antibody blockade (anti-TGF-b and anti-PD-L1) led to a significant increase in the number of cytotoxic CD8 + T cells in the TME. The co-inhibition of TGF-b and PD-L1 converted an immune excluded tumor phenotype to an inflamed phenotype, supporting the fact that TGF-b signaling prevents T-cell invasion. T cell localization was not affected with either antibody as monotherapy. Thus, TAM expressing immune checkpoint receptor ligands limit the functions of effector T cells, NK cells, and dendritic cells, and attenuate the effects of immune checkpoint inhibitor therapy.
To prevent phagocytosis, upregulated CD47 surface proteins on tumor cells provide a "do not eat me" signal by ligating the inhibitory TAM-receptor signal regulatory protein alpha (SIRPa). As CD47 also promotes the proliferation of cancer cells via the PI3K/AKT pathway, the CD47 signaling pathway is considered an important mechanism of therapy resistance. Inhibition of CD47 could be a promising therapeutic strategy, particularly in combination with immune checkpoint inhibitors.
In mouse models of melanoma, colon carcinoma, and lymphoma, dual targeting of CD47 and PD-L1 was found to enhance anti-tumor effectsand several clinical trials evaluating the efficacy of CD47 or SIRPa monoclonal antibodies as monotherapy or in combination with immune checkpoint inhibitors are underway; ClinicalTrials.gov).
As VEGF-A is overexpressed in both tumor cells, CAF, and TAM and is associated with cancer progression and dissemination, it represents the main target of anti-angiogenic drugs in cancer therapy. These drugs are widely used in the treatment of various cancers and have resulted in increased overall survival or progression-free survival in gynecologic cancers, CRC, and gastric cancer. However, due to antiangiogenic drug resistance of tumor cells, metastasis and mortality continue to occur during and after cessation of treatment. This resistance comprises the amplification of pro-angiogenic genes, secretion of multiple proangiogenic factors, and recruitment of proangiogenic BMDSC. Bevacizumab, a humanized monoclonal antibody that targets all VEGF-A isoforms and the first anti-angiogenic drug approved for clinical application, is efficacious in various malignancies such as CRC and glioblastoma. Today, most clinical studies use antiangiogenetic drugs in combinatory regimens, e.g. lenvatinib (multiple kinase inhibitor) inhibiting both VEGFR 1-3 and PDGFR and Pembrolizumab (anti-PD-1) for the treatment of endometrial cancer (NCT03517449).
The anti-diabetic drug metformin appears to be a promising therapeutic agent in neoadjuvant and adjuvant settings. The metformin-induced antitumor and anti-angiogenic effects are partly related to the skewing of TAM polarization from M2-to M1-like phenotype and significant inhibition of tumor angiogenesis. Currently, there is very little insight into the mechanism through which metformin modulates macrophage function. However, an in vitro study on breast cancer cells and TAM polarization revealed that metformin treatment activated AMPK-NF-kB signaling in cancer cells. These molecules participate in the regulation of M1 and M2 inducing cytokines. Metformin was observed to increase macrophage expression of M1-related cytokines IL-12 and TNF-a and attenuate the expression of the M2-related cytokines IL-8, IL-10, and TGF-b. Furthermore, the secretion of important cytokines for the M2 phenotype (e.g. IL-4, IL-10, and IL-13) was inhibited in metformin-treated cancer cells.
In cultures of human cholangiocarcinoma cells, and at concentrations corresponding to plasma levels of metformin in diabetic patients, metformin inhibited proliferation and cell migration and induced apoptosis. Expression of vimentin (mesenchymal marker) and EMT genes was downregulated and expression of cytokeratin-19 (epithelial marker) was upregulated. The findings from the multiple ongoing trials (173) may convey a deeper understanding of the antitumor function of metformin in the near future.
# Discussion
In a solid tumor, the balance between growth and differentiation is determined by the TME. TAM and CAF promote cancer evolution through the inflammatory, immunosuppressive, angiogenic, energy-rich environment, and also suppress cancer cells via predominantly unknown mechanisms. The presence and precise functions of CAF and TAM in the TME are extremely complex and incompletely understood, and only a few studies describe the interplay between these cells. The general perception is that the TME strongly modulates tumor cells through all phases of disease progression, and as each tumor is comprised of multiple clones with myriads of cell types and signaling molecules, the heterogeneity of each tumor may therefore require unique therapeutic approaches. Improving our understanding of the TME including the impact of stromal cells, immune cells, and ECM components, is vital for the innovation of therapeutic strategies. Hence, cellcell communication within the TME should be integrated into future cancer research. However, the manipulation of the immune system and/or stromal components within the TME during cancer treatment can be unpredictable. The regulation/ eradication of a-SMA + or FAP + CAF have had variable results and currently, targeting CAF or TAM individually does not seem to be an appropriate approach.
A CAF-directed therapy could be designed against specific pro-tumorigenic factors that in turn could prevent CAF activation or CAF functions. The reprogramming of CAF back into a normal resting phenotype would be a desirable option; however, targeting FAP has had a minimal response in human trials. Drugs that target CAF may emerge as a complement to immunotherapies in solid tumors, though a major obstacle in the precision strategy of CAF-based therapy is that neither a-SMA nor FAP is exclusively expressed by CAF.
In theory, TAM antagonists could be used to overcome resistance to immunotherapy; nevertheless, the type of approach is yet to be determined. The lack of macrophage selectivity has so far hindered its introduction into the clinic.
Monoclonal antibodies blocking the interaction between CD47 on tumor cells and SIRPa on innate immune cells is another interesting direction for future research. Other potential treatment targets are the MMPs. In the TME, MMPs are expressed by various cell types, and a number of MMP inhibitors have been tested in phase 1, 2, and 3 clinical trials. Unfortunately, all trials across different cancer types and stages have failed to provide any improvements in the clinical outcomes. Nonetheless, the field is advancing fast with the development of small-molecule inhibitors and antibodies targeting specific domains of pro-tumorigenic MMPs.
In PDAC, the TME is an important contributor to tumor progression and prognosis. The increasing amount of ECM and fibrosis promote tumor progression and correlate with shorter survival. The aberrant TGF-b signaling in cancer cells leads to an increased epithelial signal transducer and activator of transcription 3 (STAT3) activity, resulting in increased ECM fibrosis. Therefore, the concept of reducing tumor aggressiveness by interfering with STAT3 hyperactivity seems intriguing.
Notably, a recent study demonstrated that increased phosphorylation of STAT3 in CAF was associated with reduced overall survival in CRC patients. To improve response rates and increase the number of responding cancer types, combination therapies using STAT3 inhibitors and immune checkpoint inhibitors are now being undertaken.
In conclusion, combinations of immune-modulating agents are gaining more and more ground in oncology. CAF and TAM hold significant potential to improve targeted therapy and outcomes in cancer treatment when combined with existing therapies. Although in its naive stages, the TME modulating technology is an active field of research that holds immense prospects for researchers, clinicians, and patients.
# Author contributions
HR: idea, design, intelectual contents, writing the manuscript, and crreation of ilustration. AO: intelectual contents, writing the manuscript, and creation of illustrations. SG: language editing, intelectual contents, manuscript structure, and creation of tables. IG: overall design, intelectual contents, and editing. All authors contributed to the article and approved the submitted version. FIGURE 3 | Interplay between tumor, stromal, and immune cells. Depicts the interplay between cancer-associated fibroblasts (CAF), tumor-associated macrophages (TAM), growth factors, cytokines, interleukins, and immune cells in the tumor microenvironment (TME). CAF are the predominant cell type in the tumor stroma, contributing to the proliferative, pro-inflammatory, immunosuppressive, angiogenic, pro-invasive and pro-metastatic TME. They secrete various growth factors including TGF-b, vascular endothelial growth factor (VEGF), fibroblast growth factor 5, growth differentiation factor 15, and hepatocyte growth factor. CAF also produce cytokines and interleukins that may have both immunosuppressive and immuno-activating effects on various leukocytes, including CD8 + T cells, immunosuppressive regulatory T cells (T regs ) and macrophages. A subset of CAFs have myofibroblasts characteristics (myCAF) and play a major role in the development of the fibrotic stroma in the TME including the regulation of collagen fibre elongation. Growth factors and immunosuppressive cytokines produced by TAM enhance motility, intravasation, and invasion of tumor cells, while stimulating angiogenesis and suppressing T cell infiltration. Additionally, TGF-b produced by TAMs activate immunosuppressive T regs.
## Raskov et al.
Cancer-Associated Fibroblasts and Tumor-Associated Macrophages |
Umbilical cord-care practices in low- and middle-income countries: a systematic review
Background: Neonatal sepsis is the third leading cause of deaths for infants in their first month of life. The newly cut umbilical cord can be a pathway for bacteria that can cause newborn sepsis and death. Optimal umbilical cord care practices for newborns and during the first week of life, especially in settings with poor hygiene, has the potential to avoid these preventable neonatal deaths. The purpose of this review of cord care practices is to assist in the development of behavior-change strategies to support introduction of novel cord-care regimens, particularly 7.1% chlorhexidine digluconate for umbilical cord care. Methods: We searched domestic and international databases for articles that were published in English between January 1, 2000, and August 24, 2016. We found 321 articles and reviewed 65 full-text articles using standardized inclusion criteria. The primary criteria for inclusion was a description of substances applied to the umbilical cord stump in the days following birth. Results: We included 46 articles in this review of umbilical cord-care practices. Articles included data from 15 lowand middle-income countries in sub-Saharan Africa (8 countries), Asia (5 countries), North Africa (1 country), and Latin America and the Caribbean (1 country). Findings from this review suggest that documentation of cord-care practices is not consistent throughout low-and middle-income countries, yet existing literature depicts a firm tradition of umbilical cord care in every culture. Cord-care practices vary by country and by regions or cultural groups within a country and employ a wide range of substances. The desire to promote healing and hasten cord separation are the underlying beliefs related to application of substances to the umbilical cord. The frequency of application of the substance (either the number of days or the number of times per day the substance was applied), and source and cost of products used is not well-characterized. Conclusions: This desire to actively care for the umbilical cord of a newborn-as noted in the variety of cord care practices and beliefs identified in this review-points toward the need to contextualize any behavior change approach to align with the local culture.
# Background
Neonatal sepsis is responsible for more than 15% of neonatal deaths worldwide [bib_ref] Global, regional, and national causes of child mortality in 2000-13, with projections..., Liu [/bib_ref] and is the third leading cause of deaths for infants in their first month of life. A newly cut umbilical cord can be a pathway for bacteria that can cause newborn sepsis and death.. Optimal umbilical cord care practices for newborns and during the first week of life, especially in settings with poor hygiene, has the potential to avoid these preventable neonatal deaths.
Harmful traditional cord-care practices are often cited as an important public health concern [bib_ref] Ghee applications to the umbilical cord: a risk factor for neonatal tetanus, Traverso [/bib_ref] [bib_ref] Cow dung, rock salt, and medical innovation in the Hindu Kush of..., Mull [/bib_ref]. A clear understanding of behavioral intention underlying traditional cord care practices in low-and middle-income countries can be helpful in addressing high rates of neonatal sepsis. Although systematic evidence reviews of cord-cleansing practices have been conducted previously [bib_ref] Topical umbilical cord care for prevention of infection and neonatal mortality, Karumbi [/bib_ref] [bib_ref] Umbilical cord antiseptics for preventing sepsis and death among newborns, Imdad [/bib_ref] , the qualitative nature of cord-care practices has not been summarized to-date. This review fills a gap in the literature by systematically reviewing available evidence related to traditional cord-care practices and assessing the likely impact of product categories on infection risk.
# Methods
Our initial search focused on studies that described traditional umbilical cord care practices globally. For the purposes of this article, traditional practices were those that focused on the cultural beliefs and customs that guided how the umbilical cord was cared for, including the length of the cord stump, substances applied, and the decisions regarding disposal of the cord stump. We developed systematic searches for PubMed and Google Scholar using controlled vocabulary (Additional file 1: Search Terminology). Initial criteria for eligibility were determined by topic, time period, and language of the publication. We included articles that were published between January 1, 2000, and January 30, 2016. A second search was performed in August 2016 to account for any publications during the intervening months. The language of publication was limited to English. References in the identified articles were reviewed to determine if other sources would be pertinent and additional articles were abstracted if relevant.
The original search yielded 321 articles, from which 107 duplicates were excluded. A reviewer then screened titles and abstracts of the remaining 214 articles to determine suitability for inclusion. Articles that did not meet the criteria included those unrelated to the application of substances to the umbilical cord, articles focused on clinical trials comparing a variety of antiseptic applications to the umbilical cord, articles wherein the authors related only secondary data sources regarding umbilical cord-care practices, and articles that were unrelated to umbilical cord care, but had appeared in the search due to a common term, such as spinal cord. A total of 65 full-text articles were then reviewed using standardized inclusion criteria. The primary criterion for inclusion was a description of substances applied to the umbilical cord stump in the days following birth. Based on these criteria, a total of 46 of the 65 articles were included in this review. Secondary data about beliefs in relation to umbilical cord care and other cord-care practices were also recorded, if available. Data regarding cord-care practices were extracted from the articles using a standardized tracking form in Excel. Data items included:
○ What was used to cut the umbilical cord? ○ What was used to tie the umbilical cord? ○ Applications of a substance to the umbilical cord stump.
○ What substance was applied? ○ How often it was applied? ○ How many days it was applied? ○ Why was it applied (belief )? ○ Who applied the substance? ○ Source of product supply. ○ Cost of product applied. ○ Other newborn skin-care practices, such as infant massage, that could contribute to the development of neonatal sepsis or tetanus were also tracked.
We synthesized data relating to cord-care practices and substance used on the cord by country. Because most of the studies were qualitative or observational in design, we were unable to draw any statistical comparisons. This review reporting followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (known as PRISMA) reporting guidelines, as warranted.
# Results
A total of 46 articles were included in this review of umbilical cord-care practices. [fig_ref] Figure 1: PRISMA flow diagram for this review article [/fig_ref] presents the flow diagram of the review process.
The 46 articles included data from 15 distinct lowand middle-income countries in sub-Saharan Africa (8 countries), Asia (5 countries), North Africa (1 country), and Latin America and the Caribbean (1 country). In sub-Saharan Africa, the majority of articles came from Uganda (6), followed by Tanzania (4), Ethiopia and Nigeria (3 each), Ghana and Zambia (2 each), and Benin and Sierra Leone (1 each). In Asia, the majority of articles came from Pakistan (7), followed by India and Nepal (5 each), Bangladesh (3), and Turkey (2). In North Africa, one article came from Egypt. In the Latin America/Caribbean region, one article came from Haiti. [fig_ref] Table 1: Articles included in this review, with study details [/fig_ref] provides general details about the articles included in this review. While country income classification was not a predetermined criteria for inclusion or exclusion, umbilical cord care articles from high-income countries were solely focused on the comparisons and uses of antiseptics and; therefore, none are included in this review based on the exclusion criteria.
## Beliefs
Beliefs related to the application of substances to the umbilical cord varies by country and by regions or cultural groups within a country. The intention behind applying a substance to the umbilical cord is to promote healing [bib_ref] Hurdles and opportunities for newborn care in rural Uganda, Byaruhanga [/bib_ref] [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref] [bib_ref] Newborn care practices among mother-infant dyads in urban Uganda, Kayom [/bib_ref] [bib_ref] A cross sectional study of newborn care practices in Gilgit, Pakistan, Khan [/bib_ref] and hasten the separation of the cord [bib_ref] Hurdles and opportunities for newborn care in rural Uganda, Byaruhanga [/bib_ref] [bib_ref] Determinants of cord care practices among mothers in Benin City, Abhulimhen-Iyoha [/bib_ref] [bib_ref] Traditional neonatal care practices in Turkey, Alparslan [/bib_ref] [bib_ref] Umbilical cord care in Ethiopia and implications for behavioral change: a qualitative..., Amare [/bib_ref] either by keeping the cord stump moist [bib_ref] A cross sectional study of newborn care practices in Gilgit, Pakistan, Khan [/bib_ref] [bib_ref] Umbilical cord care in Ethiopia and implications for behavioral change: a qualitative..., Amare [/bib_ref] [bib_ref] Improving hygiene in home deliveries in rural Ghana: how to build on..., Hill [/bib_ref] or by drying it out [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref] [bib_ref] Local understandings of care during delivery and postnatal period to inform home..., Degefie [/bib_ref] [bib_ref] Newborn care practices among slum dwellers in Dhaka, Bangladesh: a quantitative and..., Moran [/bib_ref] [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref] [bib_ref] Clean home-delivery in rural Southern Tanzania: barriers, influencers, and facilitators, Shamba [/bib_ref] [bib_ref] Acceptability of evidence-based neonatal care practices in rural Uganda -implications for programming, Waiswa [/bib_ref] to prevent pain/infection/bleeding [bib_ref] A cross sectional study of newborn care practices in Gilgit, Pakistan, Khan [/bib_ref] [bib_ref] Determinants of cord care practices among mothers in Benin City, Abhulimhen-Iyoha [/bib_ref] [bib_ref] Local understandings of care during delivery and postnatal period to inform home..., Degefie [/bib_ref] [bib_ref] Newborn care practices among slum dwellers in Dhaka, Bangladesh: a quantitative and..., Moran [/bib_ref] [bib_ref] Clean delivery practices in rural northern Ghana: a qualitative study of community..., Moyer [/bib_ref] , or to keep the "wind" (evil spirits) or cold/air [bib_ref] A cross sectional study of newborn care practices in Gilgit, Pakistan, Khan [/bib_ref] [bib_ref] Local understandings of care during delivery and postnatal period to inform home..., Degefie [/bib_ref] [bib_ref] Newborn cord care practices in Haiti, Walsh [/bib_ref] out of the infant. [fig_ref] Table 2: Articles included in this review, with cord care practicesAlparslan [13] Jpn J... [/fig_ref] provides an overview of cord-care practices described in each article included in this review. [fig_ref] Table 3: Substances used, by category [/fig_ref] illustrates the types of substances applied to the cord by country. The substance applied may depend upon the perceived nature of the cord. A moisturizing substance is applied if the cord is too brittle and a drying substance is applied if the cord takes too long to separate. For example, in southern Zambia, petroleum jelly or mabono (wild fruit) oil might be used if the cord is cracking or bleeding, while charcoal dust, baby powder, or burnt pumpkin stem may be used if the cord takes too long to separate [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref]. In areas of Tanzania, Uganda, and Zambia [bib_ref] Hurdles and opportunities for newborn care in rural Uganda, Byaruhanga [/bib_ref] [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref] [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref] [bib_ref] Acceptability of evidence-based neonatal care practices in rural Uganda -implications for programming, Waiswa [/bib_ref] [bib_ref] Understanding home-based neonatal care practice in rural southern Tanzania, Mrisho [/bib_ref] , the infant cannot leave the home until the cord separates and/or the mother cannot return to her chores until that time. In the Tonkolili district of Sierra Leone, a traditional birth attendant noted the purpose of applying pounded cassava to the cord: "It will help the umbilical recover easily and the child will walk fast".
The substances applied may also vary by the infant's perceived gestational age or state of health. For example, in the Choma District in Zambia, the substance applied to a newborn's umbilical cord differs by the newborn's perceived gestational age, as substances used for fullterm babies are considered too strong for preterm babies. A black powder made from the burnt stem of the pumpkin plant is applied to the umbilical cord of fullterm infants while a green powder made from the dried roots of the mweeye plant is applied to the cord of preterm/small infants as it is considered to be gentler than the black powder [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref]. Also in southern Zambia, substances such as petroleum jelly, mabono (wild fruit) oil, cooking/motor oil, charcoal, dried cow dung or chicken droppings, burnt pumpkin stem, and crushed loma (wasps' nest) are applied to the cord of the healthy infant. However, a separate set of substances that are considered to be medicinal are applied if the cord is red or if pus appears. The substances that are considered to be medicinal include: python oil, breast milk, alcohol, banana, cow dung, mukunku (bark of a tree), traditional herbs, or dirt from a pounding stick [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref].
The length of the cord may have specific importance. In southern Zambia, it is believed that if the cord is too long, it will take too long to heal, and if it is too short, "the air will go in and this will make the baby die" [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref]. Substances applied to the stump vary by whether child is classified (at birth) as "sick" or "healthy". If there are blood clots in the umbilical cord, a sign of abnormality and future illness, the child is taken to a healer who treats the stump with herbs or the child is taken to a hospital. Until the cord separates, mother is expected to apply a wrap around the infant's waist so that the cord does not touch the genitals (will make them infertile as an adult) and the cord is not supposed to fall on to the floor. Once the cord separates, the baby is bathed in cold water to make them strong.
Breast milk dripped onto cord. Full-term healthy babies: black powder (made from burn stem of pumpkin). Pre-term babies: green powder (ground dried roots of mweeye plant). This is considered gentler for preterm babies.
If the appropriate herbs cannot be found then brick ash is used. Less common: after the cord falls off, fresh dried chicken dung is mashed and put onto the wound (rooster for male babies, hen for female babies). Participants explained that this tradition originated in Zimbabwe.
Shamba [bib_ref] Clean home-delivery in rural Southern Tanzania: barriers, influencers, and facilitators, Shamba [/bib_ref] J Health Popul Nutr. 2013 Tanzania A study to explore childbirth-related hygiene and newborn care practices in community deliveries in southern Tanzania. New blades were used in almost all instance, but new cord ties were rarely used. Substance was applied to cord by many mothers.
Breast milk, talcum powder, oil, petroleum jelly, ash, and dirt.
SharkeyHealth Policy Plan. 2016 Sierra Leone A study to understand maternal and neonatal care practices in four rural districts of Sierra Leone.
Pounded cassava.
Sharma [bib_ref] Dirty and 40 days in the wilderness: eliciting childbirth and postnatal cultural..., Sharma [/bib_ref] BMC Pregnancy Childbirth. 2016 Nepal A study to understand beliefs around childbirth and postnatal card in rural Nepal, particularly focusing on beliefs related to women being polluted following childbirth. A hasiya (scythe), a sickle, or a razor blade may be used to cut the umbilical cord and then a substance may be applied.
Antiseptic, cooking oil, ghee, plain water, toothpaste, and ash. The length of the cord is also believed to indicate the length of the genitals in both men and women [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref]. Further, in southern Zambia and on the island of Pemba in the Zanzibar archipelago, Tanzania, the umbilical cord may be wrapped or bound with a strip of cloth to prevent it from touching the groin area [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref] [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref] [bib_ref] Delivery, immediate newborn and cord care practices in Pemba Tanzania: a qualitative..., Dhingra [/bib_ref] , which is sometimes believed to cause infertility as an adult [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref]. In areas of Zambia and Tanzania, where it is important to protect the cord from being taken by someone who wishes the infant or the family ill, the cord, and often the placenta, are disposed of through burial in a sacred place or by burning, or are sometimes put in a pit latrine/toilet to prevent them from being unearthed [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref] [bib_ref] Understanding home-based neonatal care practice in rural southern Tanzania, Mrisho [/bib_ref]. In southeastern Turkey, the cord may be buried in a special place to help define the child's path in life, such as burying the cord at a mosque or a school to help the child grow up to be a religious or educated person, respectively [bib_ref] Traditional postpartum practices of women and infants and the factors influencing such..., Geçkil [/bib_ref].
## Frequency of application
Few articles reported on the frequency of application of the substance (either the number of days or the number of times per day the substance was applied). In Ethiopia, the substance is applied one to three times per day up to the seventh day of life. However, the applications may not begin until the newborn is two or three days old [bib_ref] Umbilical cord care in Ethiopia and implications for behavioral change: a qualitative..., Amare [/bib_ref]. In the Brong Ahafo region of Ghana, a substance is applied anywhere from every 30 min to 3 times per day [bib_ref] Improving hygiene in home deliveries in rural Ghana: how to build on..., Hill [/bib_ref]. In urban Uganda, mothers reported cleaning the cord with a substance at least twice per day [bib_ref] Newborn care practices among mother-infant dyads in urban Uganda, Kayom [/bib_ref]. On Pemba Island in the Zanzibar archipelago, Tanzania, multiple substances were applied to the umbilical area beginning on the sixth day after birth. In this study of more than 1000 infants born at home, only 10% (n = 109) had a substance applied to the umbilical cord and less than 11% of those applications were made in the first 48 h of life [bib_ref] Incidence and risk factors for newborn umbilical cord infections on Pemba Island, Mullany [/bib_ref]. A study in Sylhet District in Bangladesh, involving 39 in-depth interviews of mothers, fathers, grandmothers, and traditional birth attendants and data from more than 6000 household surveys of mothers, found that substances were applied until the cord separated and that either turmeric or ginger are applied at birth and then a combination of mustard oil and garlic are applied twice daily until the separation of the cord [bib_ref] Local understandings of vulnerability and protection during the neonatal period in Sylhet..., Winch [/bib_ref]. Few studies identified who usually applied the substance to the cord. In the 19 studies where it was reported, either the mother or grandmother of the infant or a senior woman [bib_ref] Improving hygiene in home deliveries in rural Ghana: how to build on..., Hill [/bib_ref] [bib_ref] Clean home-delivery in rural Southern Tanzania: barriers, influencers, and facilitators, Shamba [/bib_ref] [bib_ref] Household knowledge and practices of newborn and maternal health in Haripur district, Khadduri [/bib_ref] in the household applied the substance to the cord. In only a very few cases did a traditional birth attendant or health worker [bib_ref] Umbilical cord care in Ethiopia and implications for behavioral change: a qualitative..., Amare [/bib_ref] [bib_ref] Clean home-delivery in rural Southern Tanzania: barriers, influencers, and facilitators, Shamba [/bib_ref] apply a substance to the cord.
## Cost and source of substance
Few articles investigated the source of the substances applied to the umbilical cord and none reported on the cost of the substances applied. Shea butter or cooking oil were purchased from the market [bib_ref] Improving hygiene in home deliveries in rural Ghana: how to build on..., Hill [/bib_ref] [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref]. One study reported that the cooking oil applied to the cord is generally a recycled product bought at the local market that had been previously used [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref].
# Discussion
This study presents a view of traditional cord-care practices as reported during the last fifteen years in low-and middle-income countries. The desire to care for the umbilical cord of an infant appears to be universal in all cultures. Of interest is the description of the range of products applied to the newly cut cord and that substances are applied to the umbilical cord to infants born at home as well as in facilities. Participants in studies from Petit-Gôave, Haiti, and Karachi, Pakistan, reported that if an infant was born in a health facility, a substance would be applied to the umbilical cord upon returning home [bib_ref] Newborn cord care practices in Haiti, Walsh [/bib_ref] [bib_ref] Newborn care knowledge and practices among mothers attending pediatric outpatient clinic of..., Gul [/bib_ref]. The possible harm of these substances has not been fully quantified. In some cases, such as with kohl and surma used in Egypt and Pakistan, the lead and antimony included in the product is most likely harmful. Anecdotal literature often refers to the use of dung as a harmful traditional practice. In contrast, we found use of chicken/lizard/cow dung reported in only three countries (Haiti, Uganda, and Zambia), thereby suggesting that this practice is not as widespread as depicted or that it, as reported by a traditional birth attendant in northern Ghana, was practiced in the past and has since ceased [bib_ref] Clean delivery practices in rural northern Ghana: a qualitative study of community..., Moyer [/bib_ref]. It is also possible that application of preparations using dung is underreported in published literature. For example, unpublished formative research from the Kenieba and Koutiala Districts of Mali report use of bassa bo (lizard excrement powder mixed with shea butter) and bagani dji (insect powder mixed with shea [fig_ref] Table 3: Substances used, by category [/fig_ref] Substances used, by category (Continued) Antiseptics (unspecified) [bib_ref] Incidence and risk factors for newborn umbilical cord infections on Pemba Island, Mullany [/bib_ref] [bib_ref] Dirty and 40 days in the wilderness: eliciting childbirth and postnatal cultural..., Sharma [/bib_ref] [bib_ref] Neonatal home care practices in rural Egypt during the first week of..., Darmstadt [/bib_ref] [bib_ref] Potential role of traditional birth attendants in neonatal healthcare in rural southern..., Falle [/bib_ref] [bib_ref] Intra-and inter-household differences in antenatal care, delivery practices and postnatal care between..., Ghosh [/bib_ref] [bib_ref] Home care practices for newborns in rural southern Nepal during the first..., Karas [/bib_ref] [bib_ref] Risk factors for umbilical cord infection among newborns of southern Nepal, Mullany [/bib_ref] [bib_ref] Home delivery and newborn care practices among urban women in western Nepal:..., Sreeramareddy [/bib_ref] Egypt, India, Nepal, Tanzania
Baby powder/talcum powder [bib_ref] Local perceptions, cultural beliefs and practices that shape umbilical cord care: a..., Herlihy [/bib_ref] [bib_ref] Newborn care practices among slum dwellers in Dhaka, Bangladesh: a quantitative and..., Moran [/bib_ref] [bib_ref] Clean home-delivery in rural Southern Tanzania: barriers, influencers, and facilitators, Shamba [/bib_ref] [bib_ref] Acceptability of evidence-based neonatal care practices in rural Uganda -implications for programming, Waiswa [/bib_ref] [bib_ref] Delivery, immediate newborn and cord care practices in Pemba Tanzania: a qualitative..., Dhingra [/bib_ref] [bib_ref] Incidence and risk factors for newborn umbilical cord infections on Pemba Island, Mullany [/bib_ref] [bib_ref] Breast feeding practices and newborn care in rural areas: a descriptive cross-sectional..., Madhu [/bib_ref] [bib_ref] Neonatal mortality and prevalence of practices for newborn care in a squatter..., Ayaz [/bib_ref] [bib_ref] Effect of village health team home visits and mobile phone consultations on..., Ayiasi [/bib_ref] Other product categories (i.e., oils, herbs/spices/plants, mineral/powder, water, bodily fluids, food, personal care/ medical products) and processes such as heat treatment may or may not be harmful and could warrant further investigation. Neonates and young infants are more prone to infection than older children and adults and evidence suggests that their immune systems are still developing rather than being fully formed at the time of birth. Further, newborns appear to be particularly vulnerable to the types of intracellular pathogens that commonly cause neonatal sepsis [bib_ref] Neonatal innate immunity to infectious agents, Maródi [/bib_ref] , such as some strains of streptococcus, Escherichia coli, and Listeria monocytogenes.
Concern has been expressed about transmission of HIV from mother to child through the application of breast milk to the umbilical cord [bib_ref] Skin, thermal and umbilical cord care practices for neonates in southern, rural..., Sacks [/bib_ref]. Also, the use of numbati (water that has been used to wash an adult woman's genitals) in Tanzania [bib_ref] Understanding home-based neonatal care practice in rural southern Tanzania, Mrisho [/bib_ref] , could potentially pose a risk for transmission of HIV or other diseases. The application of unhygienic substances to or around the umbilical cord stump has been linked to tetanus in infants [bib_ref] Ghee applications to the umbilical cord: a risk factor for neonatal tetanus, Traverso [/bib_ref] [bib_ref] Traditional neonatal care practices in Turkey, Alparslan [/bib_ref] [bib_ref] Traditional postpartum practices of women and infants and the factors influencing such..., Geçkil [/bib_ref] [bib_ref] A matched case-control study of risk factors for neonatal tetanus in Karachi,..., Raza [/bib_ref] [bib_ref] A review of neonatal tetanus in University of Maiduguri Teaching Hospital, Northeastern..., Alhaji [/bib_ref] [bib_ref] Maternal and neonatal tetanus, Roper [/bib_ref] [bib_ref] Effect on neonatal tetanus mortality after a culturally-based health promotion programme, Meegan [/bib_ref]. Clostridium tetani bacteria, found in soil, dust, saliva, animal dung, and other sources, are the cause of tetanus infection.
Some substances, which are known to be harmful in other contexts, pose an unclear risk when applied to the umbilical cord. For example, motor/machine oil contains high levels of chemical additives and, once used for its intended purpose, can contain high levels of heavy metals and other minerals. While used motor/ machine oil is well known to be harmful to the environment and causes dermatitis through long-term exposure, the health risks due to short-term exposure through an open wound is not well defined. Boric acid powder is used as a pesticide and has caused seizures and death in infants when ingested , but also has known antifungal properties and has been used as a treatment for conditions such yeast infections . However, sources differ on whether it is safe to use boric acid on open wounds. Pyriproxyfen powder, a pesticide, poses minimal risk to humans in small quantities. Other substances, such as oils, herbs, plants, food products, and heat treatments applied to the cord may be harmful depending on whether they have been contaminated in some way, such as through unhygienic preparation or storage.
This critical information regarding traditional cordcare practices can serve as the stepping stone to behavior change. We focused our review on a comprehensive description of reported cord-care practices with the intention of employing the knowledge to develop behavior-change strategies to support introduction of novel cord-care regimens, particularly 7.1% chlorhexidine digluconate for umbilical cord care. Randomized controlled trials investigating the use of 7.1% chlorhexidine digluconate for umbilical cord care have been conducted in Nepal [bib_ref] Topical applications of chlorhexidine to the umbilical cord for prevention of omphalitis..., Mullany [/bib_ref] , Bangladesh [bib_ref] The effect of cord cleansing with chlorhexidine on neonatal mortality in rural..., Arifeen [/bib_ref] , and Pakistan [bib_ref] Topical application of chlorhexidine to neonatal umbilical cords for prevention of omphalitis..., Soofi [/bib_ref]. A meta-analysis of the three studies demonstrated that application of chlorhexidine to the umbilical cord of the newborn led to a 23% reduction in all-cause neonatal mortality and a reduction in omphalitis ranging from 27 to 56% compared to control group depending on severity of infection [bib_ref] The effect of umbilical cord cleansing with chlorhexidine on omphalitis and neonatal..., Imdad [/bib_ref]. Further, the World Health Organization recommends the application of 7.1% chlorhexidine digluconate (gel or solution) to the umbilical cord of neonates who are born at home in settings with high neonatal mortality or to replace the use of a harmful, traditional substances. Despite previously reported substantial reductions in South Asia, results from recent trials in Zambia [bib_ref] Effectiveness of 4% chlorhexidine umbilical cord care on neonatal mortality in Southern..., Semrau [/bib_ref] and Tanzania [bib_ref] Efficacy of chlorhexidine application to umbilical cord on neonatal mortality in Pemba,..., Sazawal [/bib_ref] show that application of 7.1% chlorhexidine to the umbilical cord did not significantly reduce neonatal mortality rates in the study sites. This suggests that programmatic context and level of risk in the population as well as cord care practices must be considered in any behavior change initiative.
Findings from this review suggest that documentation of cord-care practices is not consistent throughout lowand middle-income countries. Given the heterogeneity of practices described in the literature, it is not clear if data from one country in a region also pertains to other surrounding countries and/or nearby ethnic groups. Overall, however, existing literature depicts a firm tradition of umbilical cord care in every culture. The desire to take some kind of action to address the newly cut umbilical cord seems to be a strong human desire. Participants in several studies noted that adhering to dry cord care was very difficult. For example, in Ghana, there was a belief that applying nothing to the cord would delay separation, cause discomfort, and potentially cause death for the infant by preventing the sore from healing and causing a sickness in the stomach [bib_ref] Improving hygiene in home deliveries in rural Ghana: how to build on..., Hill [/bib_ref]. In Uganda, the practice of dry cord care was noted as difficult to follow as it delays the cord separation and the return of the mother to chores [bib_ref] Hurdles and opportunities for newborn care in rural Uganda, Byaruhanga [/bib_ref]. This desire to actively care for a newborn could be utilized as a trigger for behavior change. For example, as applied in the Health Belief Model, knowledge of effective cord care (i.e., product, application procedure, number of days to use, number of times/day to apply) could function as a "cue to action" to allow caretakers to act in a positive manner.
This review has limitations, as noted. Reporting is not global in nature as the review was limited to sources published in English in peer-reviewed journals. Few of the studies conducted have been indepth, qualitative assessments of umbilical cord care practices; therefore, much of the detailed information about beliefs and practices surrounding umbilical cord care comes from a few sources or countries, such as Tanzania and Zambia. This could lead to the assumption that some countries place greater importance on umbilical cord care practices than other countries where the data collection focused on a variety of newborn care practices. The paucity of data from Latin America and the Caribbean reflected in this review could stem from our decision to include only English-language material. Also, it is likely that additional anecdotal information is available in gray literature.
Additional research around cost and source of products used in cord care practices could assist programs that are targeting newborn care behavior change. Likewise, deeper ethnographic and/or qualitative investigation into the underlying meaning and significance of traditional cord-care practices could assist in formulating key messages being used to generate demand for novel prophylactic products.
# Conclusions
Findings from this review suggest that documentation of cord-care practices is not consistent throughout low-and middle-income countries, yet existing literature depicts a firm tradition of umbilical cord care in every culture studied. Cord-care practices vary by country and by regions or cultural groups within a country and employ a wide range of substances. The desire to promote healing and hasten cord separation are the underlying beliefs related to application of substances to the umbilical cord. The frequency of application of the substance (either the number of days or the number of times per day the substance was applied) and source and cost of products used are not well characterized. This desire to actively care for the umbilical cord of a newborn could be utilized to promote positive behavior change such as the introduction of 7.1% chlorhexidine digluconate for umbilical cord care. The variety in cord care practices and beliefs noted in this review however points toward the need to contextualize any behavior change approach to align with the local culture.
## Additional file
Additional file 1: Search Terminology. Exact narrative sequences used for literature search.
[fig] Figure 1: PRISMA flow diagram for this review article [/fig]
[table] Table 1: Articles included in this review, with study details [/table]
[table] Table 2: Articles included in this review, with cord care practicesAlparslan [13] Jpn J Nurs Sci. 2013 Turkey A study to identify traditional neonatal care practices applied by women ages 15 to 49. Cord cutting and tying practices were not discussed. Mothers reported putting something on the cord to make it separate faster, tying the belly with a rope, and also putting a buttered cloth over the infant's infected belly.This study reports on previous studies from Turkey which identified the following substances placed on the umbilical cord: dry coffee, olive oil, rotten tree powder, myrtle, sugared fat, hellebore, black sesame, and burnt cloth. This study did not inquire about specific substances, but did report that 10.3% of the women put a substance on the cord, 7.2% tied the belly with a rope, and 1.5% put a buttered cloth over the infant's infected belly.Amare[14] BMC Int Health Hum Rights. 2014 Ethiopia Investigation of practices and perspectives related to umbilical care in Ethiopia. A new razor blade which is sometimes boiled (22/24 mothers), cord tie varies (sewing thread, thread from kerosene stove, sisal, blanket strips, etc.). Cord is usually cut before delivery of placenta. Application of substance in most cases. Those who did not apply substance were either advised by health workers or said it was not customary (5/24 mothers). [/table]
[table] Table 3: Substances used, by category [/table]
|
Association between METTL3 gene polymorphisms and neuroblastoma susceptibility: A nine‐centre case‐control study
Neuroblastoma ranks as the most commonly seen and deadly solid tumour in infancy.The aberrant activity of m 6 A-RNA methyltransferase METTL3 is involved in human cancers. Therefore, functional genetic variants in the METTL3 gene may contribute to neuroblastoma risk. In the current nine-centre case-control study, we aimed to
analyse the association between the METTL3 gene single nucleotide polymorphisms (SNPs) and neuroblastoma susceptibility. We genotyped four METTL3 gene SNPs (rs1061026 T>G, rs1061027 C>A, rs1139130 A>G, and rs1263801 G>C) in 968 neuroblastoma patients and 1814 controls in China. We found significant associations between these SNPs and neuroblastoma risk in neither single-locus nor combined analyses. Interestingly, in the stratified analysis, we observed a significant risk association with rs1061027 AA in subgroups of children ≤ 18 months of age (adjusted OR = 1.87, 95% CI = 1.03-3.41, P = .040) and females (adjusted OR = 1.86, 95% CI = 1.07-3.24, P = .028). Overall, we identified a significant association between METTL3 gene rs1061027 C>A polymorphism and neuroblastoma risk in children ≤18 months of age and females. Our findings provide novel insights into the genetic determinants of neuroblastoma.
## | introduc ti on
Neuroblastoma is a solid childhood cancer arising from sympatho-adrenergic neuronal progenitors.It accounts for approximately 5% of all paediatric cancers, but disproportionally causes 12% cancer mortality in children.The incidence rate of neuroblastoma in America is nearly 1 out of 7000.Yet, the incidence rate of neuroblastoma in China is about 1 out of 13 000.Unlike other paediatric malignancies, neuroblastoma is characterized by high phenotypic heterogeneity.Clinical outcomes of cases vary significantly from spontaneous recovery without treatment to therapy-resistant progression.Survival rate could be achieved at least 95% in patients with a non-high-risk (low-and intermediate-risk) neuroblastoma.Conversely, only 50% of patients with high-risk neuroblastoma achieve long-term survival.Over the past decades, significant advances have been made towards understanding the determinants of neuroblastoma risk.Environmental or parental exposures were reported to predispose to neuroblastoma, but warrant more validations.Previous research has suggested that there is a strong genetic component underlying neuroblastoma susceptibility.For example, most of the familial neuroblastoma harboured mutations in genes ALKand PHOX2B.A clinical trial of an inhibitor of ALK was launched soon after the initial discovery of ALK mutations.
Moreover, researchers have unceasingly found predisposing genetic polymorphisms in sporadic neuroblastoma.To be noted, all the identified genetic variations so far only revealed a small part of the genetic landscape of this malignancy. Therefore, it would be of translational interest to determine more causal genetic risk variants for improving the prevention and prognosis of neuroblastoma.
N6-methyladenosine (m 6 A) is the most prevalent modification in RNA, especially mRNA.The m 6 A modification mainly regulates gene expression at the post-transcriptional levels by affecting mRNA stability, mRNA translation and splicing. 23 m 6 A modifications are installed by RNA methyltransferases (METTL3, METTL14 and WTAP, known as 'writers'), removed by the demethylases (FTO and ALKBH5, known as 'erasers'), and recognized by m 6 A-binding proteins (YTHDF1/2/3 and IGF2BP1, known as 'readers').Emerging evidence suggests that dysregulated m 6 A modification is tightly implicated in various diseases, especially cancers.To identify METTL3 genetic variations that confer susceptibility to neuroblastoma, we performed this multi-centre epidemiology study.
## | material s and me thods
## | sample selection
The current study is a hospital-based case-control study of neuroblastoma with participants recruited from nine hospitals in China
## | polymorphism selection and genotyping
Potentially functional SNPs in the METTL3 gene were screened out from the dbSNP database and SNPinfo software.In brief, we searched for potentially functional candidate SNPs located in the 5′-flanking region, 5′-untranslated region, 3′-untranslated region and exon of METTL3. Moreover, the included SNPs should conform to: (a) the minor allele frequency >5% for Chinese Han subjects; (b) putative functional potential SNPs, which might affect transcription activity or binding capacity of the microRNA-binding site; and (c) SNPs in low linkage disequilibrium with each other (R 2 > .8). Following these criteria, four SNPs (rs1061026 T>G, rs1061027 C>A, rs1139130 A>G, and rs1263801 G>C) in the METTL3 gene were selected for the Chinese sample genotyping. As shown in , there was no significant LD (R 2 < .8) among these four SNPs of METTL3 (R 2 = .036 between rs1061026 and rs1061027, R 2 = .009 between rs1061026 and rs1139130, R 2 = .248 between rs1061026 and rs1263801, R 2 = .453 between rs1061027 and rs1139130, R 2 = .459 between rs1061027 and rs1263801, R 2 = .387 between rs1139130 and rs1263801).
Blood samples were stored at −80°C. DNA was extracted according to standard procedure, followed by genotyping using TaqMan methodology.Laboratory personnel were blinded to case/control status. We also repeatedly genotyped 10% randomly selected sample to assess the genotyping error rate and obtained concordance rates of 100%.
## K e y w o r d s
case-control study, METTL3, neuroblastoma, polymorphism, risk
# | statistical analysis
Compliance of alleles at individual loci with the Hardy-Weinberg equilibrium (HWE) was measured in controls using a chi-square test.
Differences in selected demographic variables between cases and controls were assessed by the chi-square test. Crude or adjusted (for age and gender) odds ratios (ORs) with respective 95% confidence intervals (CIs) were obtained from logistic regression analyses for the analysis of associations between polymorphisms and neuroblastoma risk. Logistic regression analyses were adopted to obtain haplotype frequencies and distinct haplotypes, with the adjustment for gender and age.The P value level of significance was .05. We did the analyses using SAS 9.1 (SAS Institute).
## | re sults
## | associations between mettl3 snps and neuroblastoma susceptibility
The clinical characteristics of the eligible participants (968 cases and 1814 controls) were depicted in . No significant differences between cases and controls were observed with respect to age (P = .536) and gender (P = .231). The associations between the four METTL3 SNPs and neuroblastoma risk were shown in .
The P values of HWE for all SNPs were >.05 in the controls, indicating none of them departing from HWE. In the single-locus analysis, all the selected variants in the METTL3 gene showed no significant association with neuroblastoma susceptibility. Then, we analysed the combined effect of risk genotypes but still failed to detect any significant association.
# | stratification analysis
# | haplotype analysis
We further examined whether the haplotypes of the four METTL3
gene SNPs are correlated to neuroblastoma risk in an order of rs1061026, rs1061027, rs1139130 and rs1263801. As shown in , the TCGG haplotype was defined as the reference group.
We failed to detect a significant relationship between neuroblastoma risk and subjects with all the haplotypes.
## | d iscuss i on
## Ack n owled g em ents
This study was supported by grants from the Natural Science The haplotype order was rs1061026, rs1061027, rs1139130 and rs1263801. b Obtained in logistic regression models with adjustment for age and gender.
## Co n fli c t o f i nte r e s t
The authors confirm that there are no conflicts of interest.
## Auth o r co ntr i b uti o n
## Data ava i l a b i l i t y s tat e m e n t
All the data were available upon request.
## O rci d
Jinhong Zhu https://orcid.org/0000-0002-0408-3101
Jing He https://orcid.org/0000-0002-1954-2892
## R e fe r e n c e s
|
Polymerized Laminin-332 Matrix Supports Rapid and Tight Adhesion of Keratinocytes, Suppressing Cell Migration
Laminin-332 (a3ß3c2) (Lm332) supports the stable anchoring of basal keratinocytes to the epidermal basement membrane, while it functions as a motility factor for wound healing and cancer invasion. To understand these contrasting activities of Lm332, we investigated Lm332 matrices deposited by normal human keratinocytes and other Lm332-expressing cell lines. All types of the cells efficiently deposited Lm332 on the culture plates in specific patterns. On the contrary, laminins containing laminin ß1 and/or c1 chains, such as Lm511 and Lm311, were not deposited on the culture plates even if secreted into culture medium. The Lm332 deposition was not inhibited by function-blocking antibodies to the a3 and a6 integrins but was inhibited by sodium selenate, suggesting that sulfated glycosaminoglycans on cell surface, e.g. heparan sulfate proteoglycans, might be involved in the process. HEK293 cells overexpressing exogenous Lm332 (Lm332-HEK) almost exclusively deposited Lm332 on the plates. The deposited Lm332 matrix showed a mesh-like network structure as analyzed by electron microscopy, suggesting that Lm332 was highly polymerized. When biological activity was analyzed, the Lm332 matrix rather suppressed the migration of keratinocytes as compared with purified Lm332, which highly promoted the cell migration. The Lm332 matrix supported adhesion of keratinocytes much more strongly and stably than purified Lm332. Integrin a3ß1 bound to the Lm332 matrix at a three times higher level than purified Lm332. Normal keratinocytes prominently showed integrin a6ß4-containing, hemidesmosome-like structures on the Lm332 matrix but not on the purified one. These results indicate that the polymerized Lm332 matrix supports stable cell adhesion by interacting with both integrin a6ß4 and a3ß1, whereas unassembled soluble Lm332 supports cell migration.
# Introduction
The interaction of animal cells with various extracellular matrix (ECM) molecules plays critical roles in both tissue construction and regulation of cellular functions such as cell adhesion, migration, proliferation and differentiation [bib_ref] Regulation of development and differentiation by the extracellular matrix, Adams [/bib_ref] [bib_ref] Integrin signaling, Giancotti [/bib_ref]. After secretion from cells, most ECM proteins are assembled into a large and complex matrix network by self-polymerization and/or interaction with other molecules [bib_ref] Basement membranes: structure, assembly and role in tumour angiogenesis, Kalluri [/bib_ref]. Basement membrane (BM) is a thin sheet of specialized ECM, in which ECM proteins such as laminins, type IV collagen, nidogens and perlecan are assembled into a complex mesh-like membrane structure [bib_ref] Basement membranes: structure, assembly and role in tumour angiogenesis, Kalluri [/bib_ref] [bib_ref] Form and function: the laminin family of heterotrimers, Colognato [/bib_ref]. It remains uncertain how each ECM molecule is assembled into the BM structure. In the BMs of various types of tissues, laminins play major roles in regulating cellular functions. Like other ECM proteins, the biological activity of laminins can be analyzed using purified proteins. However, it seems very likely that the biological activity of assembled ECM proteins differs from that of isolated proteins [bib_ref] Taking cell-matrix adhesions to the third dimension, Cukierman [/bib_ref].
One of the laminin isoforms, laminin-332 (Lm332; previously known as laminin-5), which consists of laminin a3, ß3 and c2 chains, is a major component of BMs in the skin and other stratified squamous epithelial tissues [bib_ref] A simplified laminin nomenclature, Aumailley [/bib_ref] , and associates with integrin a6ß4 to form the stable adhesion structure hemidesmosome [bib_ref] Laminin-5 and hemidesmosomes: role of the alpha 3 chain subunit in hemidesmosome..., Baker [/bib_ref] [bib_ref] Biology and function of hemidesmosomes, Nievers [/bib_ref]. Therefore, genetic mutations of Lm332 subunits cause a severe and lethal skin blistering disease, Herlitz's junctional epidermolysis bullosa [bib_ref] A homozygous nonsense mutation in the beta 3 chain gene of laminin..., Pulkkinen [/bib_ref] [bib_ref] Herlitz's junctional epidermolysis bullosa is linked to mutations in the gene (LAMC2)..., Aberdam [/bib_ref]. In vitro, Lm332 promotes cellular adhesion, motility and scattering [bib_ref] A large cell-adhesive scatter factor secreted by human gastric carcinoma cells, Miyazaki [/bib_ref] [bib_ref] Marked stimulation of cell adhesion and motility by ladsin, a laminin-like scatter..., Kikkawa [/bib_ref] [bib_ref] Kalinin is more efficient than laminin in promoting adhesion of primary keratinocytes..., Rousselle [/bib_ref]. These activities are mainly mediated through the interaction of the C-terminal laminin globular (LG) domain of the a3 chain, especially the LG3 domain with integrins a3ß1, a6ß1 and a6ß4 [bib_ref] Structural requirement of carboxyl-terminal globular domains of laminin alpha 3 chain for..., Hirosaki [/bib_ref] [bib_ref] Differential regulation of cellular adhesion and migration by recombinant laminin-5 forms with..., Kariya [/bib_ref]. Lm332 has unique activity that even in a soluble form, it induces cell migration and scattering via PKC, phosphatidylinositol 3-kinase (PI3K) and ERK activation by binding to integrins a3ß1 and a6ß1 on apical cell surface [bib_ref] The basement membrane protein laminin-5 acts as a soluble cell motility factor, Kariya [/bib_ref]. In vivo, expression of Lm332 is induced at the wounded edge of epidermis and at the leading edge of invading carcinomas. Therefore, Lm332 is thought to contribute to cell migration in wound healing [bib_ref] Cloning of the LamA3 gene encoding the alpha 3 chain of the..., Ryan [/bib_ref] [bib_ref] Deposition of laminin 5 in epidermal wounds regulates integrin signaling and adhesion, Nguyen [/bib_ref] and tumor invasion [bib_ref] Laminin-5 is a marker of invading cancer cells in some human carcinomas..., Pyke [/bib_ref] [bib_ref] Laminin-5 in the progression of carcinomas, Lohi [/bib_ref]. Indeed, keratinocytes deficient in Lm332 expression show defects in their cell migration [bib_ref] Deposition of laminin 5 in epidermal wounds regulates integrin signaling and adhesion, Nguyen [/bib_ref] [bib_ref] Bisecting GlcNAc residues on laminin-332 down-regulate galectin-3-dependent keratinocyte motility, Kariya [/bib_ref]. These facts suggest a crucial role of Lm332 in the migration of normal keratinocytes as well as invading cancer cells. In contrast, there is a report showing that Lm332 inhibits keratinocyte migration in vitro [bib_ref] Laminin-5 inhibits human keratinocyte migration, O'toole [/bib_ref]. Thus, Lm332 seems to exhibit two opposite activities, stable adhesion and cell motility both in vivo and in vitro.
Proteolytic processing of Lm332 may be at least in part responsible for the contrasting activity of Lm332. There are reports showing that the cleavage of the laminin a3 chain from the precursor (190 kDa) to the mature (160 kDa) form decreases the cell migration activity of Lm332 [bib_ref] Processing of laminin-5 and its functional consequences: role of plasmin and tissue-type..., Goldfinger [/bib_ref]. However, our previous study with recombinant Lm332 showed that both cell adhesion and motility activities of Lm332 are enhanced when the a3 chain is processed to the mature form [bib_ref] Regulation of biological activity and matrix assembly of laminin-5 by COOH-terminal, LG4-5..., Tsubota [/bib_ref]. On the other hand, the proteolytic cleavage of the laminin c2 chain seems to be more important for the Lm332-mediated cell migration than that of the a3 chain. Previous studies showed that the cleavage of the c2 chain from the precursor (150 kDa) to the mature (105 kDa) form significantly increases the cell migration activity of Lm332 [bib_ref] Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5, Giannelli [/bib_ref] [bib_ref] Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5, Koshikawa [/bib_ref]. However, because the difference in the cell motility activity between the two forms of Lm332 is not very striking [bib_ref] Regulation of biological activity of laminin-5 by proteolytic processing of gamma2 chain, Ogawa [/bib_ref] , it is difficult to consider that only proteolytic processing could be responsible for the differential cell motility activity of Lm332 [bib_ref] Laminin-5 (laminin-332): Unique biological activity and role in tumor growth and invasion, Miyazaki [/bib_ref].
Many previous studies have shown that normal keartinocytes deposit Lm332 onto the surface of culture plates [bib_ref] Laminin deposition in the extracellular matrix: a complex picture emerges, Hamill [/bib_ref]. This Lm332-containing matrix may have a different biological activity from that of purified Lm332. In addition, it is important to investigate how Lm332 is deposited and assembled into the matrix after secretion. We previously established HEK293 cell lines overexpressing recombinant Lm332 (Lm332-HEK) [bib_ref] Efficient expression system of human recombinant laminin-5, Kariya [/bib_ref] or laminin-3B32 (Lm3B32-HEK) [bib_ref] Characterization of laminin 5B and NH 2 -terminal proteolytic fragment of its..., Kariya [/bib_ref]. They secrete and deposit Lm332 or Lm3B32 at a high level in culture. Using Lm332-HEK and other Lm332-expressing cell lines, we here investigated the assembly of Lm332 into matrix and its biological activity, comparing the activities of the deposited Lm332 matrix and purified Lm332 protein.
# Results
## Deposition of lm332 matrix by normal and cancer cells
Lm332 is expressed by various kinds of normal epithelial cells and cancer cells [bib_ref] A large cell-adhesive scatter factor secreted by human gastric carcinoma cells, Miyazaki [/bib_ref] [bib_ref] Differential expression of laminin-5/ladsin subunits in human tissues and cancer cell lines..., Mizushima [/bib_ref]. The C-terminal LG4-5 domain of the a3 chain and the N-terminal short arm of the c2 chain, both of which are liberated by proteolytic processing, regulate the matrix assembly and activity of Lm332 [bib_ref] The short arm of the laminin gamma2 chain plays a pivotal role..., Gagnoux-Palacios [/bib_ref] [bib_ref] Globular domains 4/5 of the laminin alpha3 chain mediate deposition of precursor..., Sigle [/bib_ref]. To characterize the Lm332-containing matrices, we first analyzed Lm332 secreted and deposited by primary normal human epidermal keratinocytes (NHK), three squamous cell carcinoma cell lines (A431, CaSki, and HSC-4) and two gastric adenocarcinoma cell lines (STKM-1 and MKN-45) by Western blotting. The relative amount of the deposited Lm332 in the ECM to the soluble one in the conditioned medium (CM) was highest in NHK cells, but all cancer cell lines deposited considerable amounts of Lm332 on the plastic surface [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , left column). In the deposited ECM, the 190-kDa precursor a3 chain, which contains the LG4-5 domain, was detected only in the cultures of NHK and STKM-1 cells. All ECMs [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , left column) and CM [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , right column) contained both the 150-kDa precursor and 105-kDa processed c2 chains, but the relative amount of the processed form to the precursor was higher in the CM than the ECMs. A nearly single band of ß3 chain was detected in all samples, indicating that this chain is relatively resistant to the proteolytic cleavage.
To see how Lm332 is deposited, the Lm332 deposition was visualized by immunofluorescent staining with the anti-a3 chain antibody BG5 in sparse cultures. In agreement with a previous study [bib_ref] Laminin 5 deposition regulates keratinocyte polarization and persistent migration, Frank [/bib_ref] , migrating NHK deposited many Lm332 signals on their trails, upper panels, Lm332). In a stationary HSC-4 cell, Lm332 was most densely observed in a perinuclear circle area and co-localized only with perinuclear actin filaments (a lower cell in, middle panels, Merged). A slowly migrating cell deposited these Lm332 proteins as spikelike or arrowhead-like spots in a semicircle area. We have previously established HEK293 cell lines which overexpress recombinant Lm332 (Lm332-HEK) [bib_ref] Efficient expression system of human recombinant laminin-5, Kariya [/bib_ref]. Slowly migrating Lm332-HEK cells produced and left different sizes of Lm332 spots uniformly behind the cells, lower panels, Lm332). However, when Lm332-HEK cells were plated at a high density, they initially deposited Lm332 in peripheral regions of individual cells, left panel), but further deposition of Lm332 covered whole surface of the culture plates with cotton-like fibers, right panel). We also examined the patterns of Lm332 matrices deposited by confluent cultures of NHK and cancer cell lines [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref]. NHK, A431 and HSC-4 produced a cloud-like or rosette-like pattern of Lm332 deposition, where small ring structures were visible especially in the matrices of NHK and A431 [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref]. Compared to these matrices, two gastric adenocarcinoma cell lines (STKM-1 and MKN-45) produced relatively homogeneous Lm332 matrix with spiny or fibrous structures [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref]. These results suggest that Lm332 is initially deposited in perinuclear or more peripheral regions, and the differences in the Lm332 deposition patterns may largely depend on the motility and cytoskeletal structure of Lm332-expressing cells. Similar Lm332 patterns were obtained when the Lm332 matrices were immunostained with antibodies recognizing the laminin a3, ß3 and c2 chains.
It has been reported that the Lm332 deposition or its assembly to ECM is mediated by cell surface molecules [bib_ref] Laminin 5 deposition regulates keratinocyte polarization and persistent migration, Frank [/bib_ref] [bib_ref] The role of alpha3beta1 integrin in determining the supramolecular organization of laminin-5..., Dehart [/bib_ref] [bib_ref] Integrin beta4 regulates migratory behavior of keratinocytes by determining laminin-332 organization, Sehgal [/bib_ref]. We . Cells were suspended in serum-free medium, inoculated at a density of 5610 6 cells per 90-mm dish, and incubated for 2 days. The resulting ECM and CM were prepared from each culture. To prepare ECM, cells were removed from the dishes by treating them with 10 mM EDTA and briefly with 20 mM NH 4 OH and then washing with PBS. The ECMs on the dishes were extracted with the SDS sample buffer. A twentieth part of the CM and a thirtieth part of the ECM were subjected to immunoblotting with the antibodies to the laminin a3, c2 and ß3 chains under reducing conditions. Bars indicate the position and size of the laminin chains. Other experimental conditions are described in ''Materials and Methods''. doi:10.1371/journal.pone.0035546.g001 examined the possible role of integrins in the Lm332 deposition. Although a mixture of function-blocking anti-integrin-a3 and anti-integrin-a6 antibodies completely blocked cell adhesion to Lm332-coated plates, it did not affect the Lm332 deposition on collagen-coated plates . On the other hand, sodium selenate, an inhibitor for the sulfation of glycosaminoglycans, inhibited the Lm332 deposition of NHK cells onto culture plates as analyzed by immunoblotting and immunocytochemistry for the laminina3 chain , under nontoxic conditions . These results strongly suggest that the Lm332 deposition is mainly mediated by cell surface sulfated glycosaminoglycans, e.g. heparan sulfate proteoglycans like syndecans, but not by integrins.
## Characterization of lm332 matrix deposited by lm332-hek cells
To characterize the Lm332-containing matrix biochemically and biologically, we used Lm332-HEK and related HEK293 cell lines, as well as purified recombinant Lm332 protein. ECMs were prepared from the cultures of Lm332-HEK [bib_ref] Efficient expression system of human recombinant laminin-5, Kariya [/bib_ref] , a3AALm332-HEK, which overexpresses an a3 chain-mutated Lm332 resistant to proteolytic processing [bib_ref] Regulation of biological activity and matrix assembly of laminin-5 by COOH-terminal, LG4-5..., Tsubota [/bib_ref] , and ß3c2-HEK, which had been transfected only with the laminin ß3 and c2 chain cDNAs [bib_ref] Efficient expression system of human recombinant laminin-5, Kariya [/bib_ref]. The ECMs and purified Lm332 were analyzed by SDS-PAGE and subsequent Coomassie Brilliant Blue (CBB) staining or immunoblotting. The CBB staining showed that Lm332-HEK [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , lane 2) and a3AALm332-HEK [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , lane 3) cell lines almost exclusively deposited the three chains of Lm332 and their proteolytic fragments. We identified two proteolytic fragments of laminin c2 chain at approximately 90-kDa (#) and 50-kDa (*). NH 2 -terminal amino acid sequencing revealed that the 90-kDa protein had the same NH 2 -terminal sequence as the mature 105-kDa c2 chain, while the 50-kDa protein was the NH 2terminal fragment separated from the 105-kDa c2 chain. These fragments were also present in the CM of Lm332-HEK cells (data not shown). Furthermore, this analysis showed that ß3c2-HEK cells secreted and deposited the ß3 and c2 chains [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , lane 4). As shown by immunoblotting [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , upper panel) as well as the CBB staining [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , a3AALm332-HEK cells deposited the 190-kDa precursor (or unprocessed) a3 chain as a major component, whereas this was never or scarcely detected in the purified Lm332 and the ECM of Lm332-HEK (named Lm332-ECM), both of which contained the 160-kDa mature (or processed) a3 chain as a major component. Immunoblotting for the c2 chain showed that the Lm332-ECM contained the 150-kDa c2 chain more than the 105-kDa processed c2 chain, but vice versa in the ECM of a3AALm332-HEK and the purified Lm332 [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , lower panel).
When the ECM was prepared from Lm332-HEK cultures 6 and 30 h after inoculation, the 190-kDa unprocessed a3 chain was found as a major band at 6 h but it was mostly converted to the 160-kDa processed form for 30 h [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref]. Similarly, the 150-kDa precursor c2 chain was found as a single band at 6 h, but it was partially converted to the 140-and 105-kDa processed forms for 30 h. These results suggest that the proteolytic cleavage of these chains mainly occurs after Lm332 is deposited and the processing of the a3 chain is much faster than that of the c2 chain.
When Lm332-ECM prepared from a 2-days culture was separated by SDS-PAGE under non-reducing conditions, bottom) were suspended in serum-free medium, inoculated at a cell density of 2610 3 cells/well on collagen-coated 8-well chamber slides and incubated for 6 h. The cultures were stained for F-actin with rhodamine phalloidin (left panels) and for Lm332 with the anti-a3 chain BG5 antibody and followed by a FITC-labeled secondary antibody (center panels), as described in ''Materials and Methods''. Right panels are merged images. In (B), Lm332-HEK cells were inoculated at a high cell density (1610 5 cells/well), incubated for 6 h (left panel) or 48 h (right panel), and stained for Lm332 as above. doi:10.1371/journal.pone.0035546.g002 . Effect of sodium selenate on Lm332 deposition by NHK cells. (A) Immunoblotting analysis. NHK cells were inoculated in serum-free medium at a density of 4610 5 cells per 35-mm dish, incubated overnight, and treated with (+) or without (2) 0.1 mM sodium selenate (Sigma) at 37uC for 24 h. After the incubation, the ECM and CM were prepared from each culture and analyzed for the laminin a3 chain by immunoblotting as described in [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] Lm332 heterotrimer could not enter the separating gel [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref]. The same was true even in the Lm332-ECM from a 6-h culture (data not shown). Since Lm332 in the ECM was separated into its three subunits by reducing SDS-PAGE [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , Lm332 was supposed to be polymerized by cross-linkage with disulfide bonds.
The density of Lm332 in the ECM deposited by Lm332-HEK cells was analyzed by Enzyme-linked immunosorbent assay (ELISA) with monoclonal antibodies against the laminin a3 and c2 chains. The Lm332 concentration on the plate was equivalent to that obtained by coating purified Lm332 at a concentration of 0.56 mg/ml and 0.61 mg/ml as analyzed for the a3 and c2 chains, respectively [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , A and B). SDS-PAGE analysis verified that this level of Lm332 was indeed present in Lm332-ECM [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] , C). The ELISA also showed that the concentration of the c2 chain in the ECM deposited by ß3c2-HEK cells was almost the same as that in the Lm332-ECM.
It is likely that ECM proteins other than Lm332 are also assembled into Lm332-ECM. We have previously reported that HEK293 cells secrete laminin-511 (Lm511) into the culture medium [bib_ref] The beta3 chain short arm of laminin-332 (laminin-5) induces matrix assembly and..., Nakashima [/bib_ref]. In immunoblotting analysis, the laminin a5 chain was detected at lower molecular sizes than the authentic a5 chain of recombinant Lm511 in the CM of Lm322-HEK cells, but it was undetectable in the ECM [fig_ref] Figure 5: Fine structures of Lm332-ECM [/fig_ref]. Similarly, the laminin ß1 and c1 chains were detected only in the CM. As shown in [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , two gastric carcinoma cell lines showed characteristic patterns of the Lm332 matrix. Therefore, we also analyzed other laminins in the CM and ECM of MKN45 cell line. Non-reducing immunoblotting analysis showed that the CM of MKN45 gastric carcinoma cells contained both Lm332 and laminin-311 (Lm311) at comparable levels, but neither the laminin ß1 nor c1 chain was detected in the ECM [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref]. These results suggest that in contrast to Lm332, laminins containing laminin ß1 and/or c1 chains, such as Lm511 and Lm311, are hardly deposited on the culture plates. Furthermore, we analyzed fibronectin, nidogen-1, perlecan, type IV collagen and type VII collagen in the ECM and CM of Lm332-HEK cells. Human fibronectin and nidogen-1 were detected in the CM but they were absent in the ECM [fig_ref] Figure 5: Fine structures of Lm332-ECM [/fig_ref] , B and C, respectively). Perlecan, type IV collagen and type VII collagen were undetectable in both ECM and CM of Lm332-HEK cells (data not shown).
Fine structure of Lm332-ECM was also analyzed by transmission electron microscopy (TEM). The TEM image of Lm332-ECM showed a mesh-like, molecular network structure covering whole surface of the glass plate, in which high density areas were distributed [fig_ref] Figure 5: Fine structures of Lm332-ECM [/fig_ref]. This suggests that Lm332 may be polymerized on the basal surface of cell membrane and transferred onto the culture substrate. Such a mesh-like structure was not found in the ß3c2-ECM, though high density areas, possibly due to protein aggregates, were distributed on the plates. Lm332coated plates showed some protein aggregates without any clear structure, suggesting that Lm332 molecules mostly detached from the plate during the staining procedure.
## Migration of keratinocytes on lm332-ecm
To show the biological activity of Lm332-ECM, we first examined effects of Lm332-ECM and purified Lm332 on migration of NHK cells. In these assays, the cell density and incubation length were minimized to neglect the effect of endogenously secreted or deposited Lm332. When the cells were plated onto culture plates pre-coated with 1.0 mg/ml Lm332, they actively and directionally migrated on the substrate [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref]. When the Lm332 concentration was increased to 2.5 mg/ml, the cell migration was reduced to a half level. To obtain regular orientation of coated Lm332 molecules, we first coated a monoclonal antibody (LSac3) that recognizes an NH 2 -terminal sequence (domain IIIa) of the a3 chain onto a plate and then bound Lm332 to the antibody-coated plate, thus allowing the integrin-binding domain LG1-3 of the a3 chain to face the cell surface receptors. This antibody-mediated Lm332coated plate supported the rapid migration of NHK cells [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref] , Ab). In contrast, NHK cells poorly migrated on both Lm332-ECM and a3AALm332-ECM [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref] , WT and AA; [fig_ref] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins [/fig_ref] , regardless of the differences in the proteolytic processing of the a3 and c2 chains as shown in [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref]. We could not measure the cell migration speed on ß3c2-ECM, because NHK cells could not fully adhere to the substrate for initial 1.5 h. However, when purified Lm332 was coated on Lm332-ECM [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref] , WT+Lm332) and ß3c2-ECM, the cell migration was significantly enhanced. These results demonstrate that coated Lm332 potently promotes migration of NHK cells, whereas Lm332-ECM rather suppresses it. The migration-suppressive activity of Lm332-ECM seemed to be independent on the orientation of Lm332 or the proteolytic processing of the a3 and c2 chains.
As described above [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , the patterns of Lm332 deposition differed depending on cell types. To confirm the suppressive effect of Lm332-ECM on cell motility, we also examined effects of the Lm332-containing ECM and the CM obtained from the culture of NHK cells. The culture plate precoated with the CM, which contained a high level of Lm332 as shown in [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , strongly promoted the migration of NHK cells [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref]. In contrast, the Lm332-containing ECM showed the suppressive effect on the cell migration. Cell migration requires morphological changes associated with dynamic actin cytoskeleton reorganization. Therefore, cell morphology and actin cytoskeleton were examined for NHK cells on purified Lm332, Lm332-ECM and ß3c2-ECM. On ß3c2-ECM, NHK cells attached but poorly spread extending filopodia-like protrusion [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref]. In contrast, NHK cells efficiently attached to Lm332-ECM and spread well showing disc-like, very flat morphology. The cells on Lm332-coated plates showed refractive morphology with typical lamellipodia at the leading edge. Such polarized cells were rarely found in the cells on Lm332-ECM or ß3c2-ECM. There was little difference in cell morphology between the Lm332 concentrations of 1.0 mg/ml and 2.5 mg/ml. Lm332-coated plates (left three columns) or deposited ECMs (right four columns), which were prepared as described below. After 1.5 h incubation, cell migration was monitored by video microscopy for 5.5 h. Purified Lm332 was coated at a concentration of 1.0 or 2.5 mg/ml on a non-treated plate, or at 1.0 mg/ml on a plate pre-coated with the anti-laminin a3 chain antibody LSac3 (Ab). ECM substrates were prepared from Lm332-HEK (WT), a3AA-Lm332-HEK (AA), and ß3c2-HEK (ßc) cultures. In the right two columns, ß3c2-ECM and Lm332-ECM were further coated with 1.0 mg/ml purified Lm332 (+Lm332) and then used for the migration assay. On ß3c2-ECM alone or a non-treated plate, NHK cells never attached and migrated during the initial 7 h (data not shown). Each bar represents the mean 6 S.D. of the migration speeds of 8 cells in each assay. These results were essentially reproduced in three independent experiments. Other experimental conditions are described in ''Materials and Methods''. (B) CM and deposited ECM (ECM) were prepared from confluent cultures of NHK cells. The CM was coated on a 24-well plate. Migration of NHK cells on these substrates was analyzed as above. (C) NHK cells were incubated for 3 h on ß3c2-ECM, Lm332-ECM and plates coated with 1.0 or 2.5 mg/ml purified Lm332, and cell morphology was observed under a phase-contrast microscope. Original magnification, 6300. Morphological characteristics of NHK cells on the different substrates were reproduced by visualization of actin cytoskeleton with rhodamine-phalloidin [fig_ref] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates [/fig_ref]. NHK cells plated on purified Lm332, regardless of its coating concentration, exhibited large lamellipodia toward the moving direction and F-actin accumulation and many retraction fibers at the rear. In contrast, the cells spread on Lm332-ECM exhibited cortical actin accumulation around the cell body and some stress fibers. The cells on ß3c2-ECM were characteristic in robust cortical actin bundles and radially extended actin filaments around cells. When NHK cells were placed on their own ECM, they showed similar morphological characteristics to the cells on Lm332-ECM. These morphological and cytoskeletal characteristics of NHK cells on the different substrates appear to reflect their motile property.
## Distinct cell adhesion activity between lm332-ecm and purified lm332
The results shown above suggested that the differential cell motility between purified Lm332 and Lm332-ECM might depend on difference in their cell-adhesive activity. This possibility was tested by incubating NHK cells at 37uC for 10 min or 40 min on each substrate [fig_ref] Figure 7: Cell adhesion activity of Lm332-ECM and purified Lm332 toward NHK cells [/fig_ref]. NHK cells could not attach to ß3c2-ECM under these conditions and even after incubation for 1.5 h. In contrast, NHK cells on Lm332-ECM attached and in parts spread at 10 min and they mostly well spread at 40 min. On a plate coated with 1.0 mg/ml Lm332, NHK cells more slowly attached and a majority of them started to spread at 40 min. Such observation was confirmed by quantitative analysis [fig_ref] Figure 7: Cell adhesion activity of Lm332-ECM and purified Lm332 toward NHK cells [/fig_ref]. Lm332-coated plates promoted cell attachment in a dose-dependent manner. The cell attachment reached the same level at 5.0 mg/ml as that to Lm332-ECM.
To examine which receptors mediate the strong adhesive activity of Lm332-ECM, inhibitory effects of various functionblocking anti-integrin antibodies and EDTA were analyzed toward NHK cells [fig_ref] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins [/fig_ref]. In agreement with previous reports [bib_ref] A homozygous nonsense mutation in the beta 3 chain gene of laminin..., Pulkkinen [/bib_ref] [bib_ref] Herlitz's junctional epidermolysis bullosa is linked to mutations in the gene (LAMC2)..., Aberdam [/bib_ref] , the cell attachment to purified Lm332 was effectively blocked by an anti-a3 integrin and weakly by an anti-ß1 integrin antibody, but anti-a2, -a5 and -a6 antibodies did not have such inhibitory effect [fig_ref] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins [/fig_ref]. A combination of anti-a3 integrin and anti-a6 integrin antibodies, as well as EDTA alone, completely blocked the cell attachment. On the other hand, the attachment of NHK cells to Lm332-ECM was scarcely blocked by any of anti-a3, -a6 and -ß1 antibodies [fig_ref] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins [/fig_ref]. However, a combination of anti-a3 and -a6 antibodies blocked the cell attachment to about 25%. The addition of anti-ß1 integrin antibody, but not anti-a2 or -a5 antibody, to the mixture of anti-a3 and -a6 antibodies completely blocked the cell attachment. When morphological effect was examined, the spreading of NHK cells on the Lm332-ECM was blocked partially by the anti-a3 integrin antibody and almost completely by the combination of anti-a3 and -a6 antibodies [fig_ref] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins [/fig_ref]. Neither anti-a6 nor anti-ß1 integrin antibody showed significant inhibition of cell spreading. ß4 integrin is known to be expressed as a6ß4, rather than a6ß1, integrin in keratinocytes [bib_ref] Integrin alpha 6/beta 4 complex is located in hemidesmosomes, suggesting a major..., Sonnenberg [/bib_ref]. Therefore, these results suggest that although NHK cells preferentially utilize integrin a3ß1 to attach to purified Lm332, integrin a6ß4 also contributes to the cell attachment to some extent. In the case of the cell attachment to Lm332-ECM, NHK cells seemed to utilize both integrins a3ß1 and a6ß4.
The results shown above suggest that the binding affinity of integrins a3ß1 and a6ß4 for Lm332-ECM may be higher than that for purified Lm332. To test this possibility, we analyzed the binding affinity of integrin a3ß1 to Lm332-ECM and purified Lm332 [fig_ref] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins [/fig_ref]. When purified integrin a3ß1 was added at varied concentrations into wells deposited with Lm332-ECM or those pre-coated with 1 mg/ml purified Lm332 in the presence of Mn 2+ , the integrin bound to the former at a much higher level than the latter. When integrin a3ß1 was added at 37 nM, the amount of integrin bound to Lm332-ECM was about 3.6-times higher than that to the coated Lm332 even though the actual concentration of Lm332 was higher in the latter wells (also see [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref].
To further confirm the strong cell adhesion activity of Lm332-ECM compared to coated Lm332, we measured cell detachment by treatment with trypsin or 10 mM EDTA. After NHK cells were allowed to adhere and fully spread on Lm332-coated plates or Lm332-ECM by incubating them for 1 h, they were treated with a diluted trypsin solution for the indicated lengths of time, followed by counting the remaining attached cells. Although the cells on purified Lm332 were almost completely detached for 10 min incubation, the majority of the cells on Lm332-ECM remained attached to the plates even after 30 min . Almost the same result was obtained when treated with EDTA alone: after 20 min incubation, 86% of NHK cells were detached from Lm332-coated plate but few cells from Lm332-ECM . These results also indicated that NHK cells firmly adhered to Lm332-ECM compared to purified Lm332.
## Hemidesmosome formation
It is well known that keratinocytes produce the stable cell adhesion structure hemidesmosome by binding to Lm332 via integrin a6ß4. The hemidesmosome structure is known to remain as insoluble spots after Triton X-100 treatment [bib_ref] Laminin 5 deposition regulates keratinocyte polarization and persistent migration, Frank [/bib_ref]. To assess the hemidesmosome formation, we analyzed localization of ß4 integrin on NHK cells by immunofluorescent staining. When NHK cells were directly subjected to the immunostaining for ß4 integrin, the cells on Lm332-ECM showed strong ring-like stain with small dot signals around nucleus, whereas those on purified Lm332 were locally stained at both front and rear edges [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , left panels). When NHK cells were immunostained after treatment with 0.5% Triton X-100, hemidesomosome-like punctuated structures of NHK cells became prominent specially at their peripheral regions on Lm332-ECM, but such peripheral staining was totally absent in the cells on purified Lm332 [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] , right panels). Based on these results, it may be concluded that NHK cells efficiently produce hemidesomosome structures containing integrin a6ß4 on Lm332 matrix but scarcely on purified Lm332.
# Discussion
In the present study, we analyzed deposition of Lm332 matrix by 7 kinds of Lm332-expressing cells including normal keratinocytes and cancer cell lines. All these kinds of cells efficiently deposited Lm332 in specific patterns onto culture plates. In a case of Lm332-HEK cells, Lm332 was an almost exclusive component in the ECM and organized into a mesh-like structure, suggesting that Lm332 was self-polymerized into the mesh structure. Furthermore, we found that the Lm332 matrix exhibited distinct activity from that of purified Lm332 protein. The former supported strong adhesion of keratinocytes but suppressed their migration as compared with the purified Lm332.
Many groups have investigated deposition and assembly of Lm332 by cultured keratinocytes [bib_ref] Regulation of biological activity and matrix assembly of laminin-5 by COOH-terminal, LG4-5..., Tsubota [/bib_ref] [bib_ref] Laminin deposition in the extracellular matrix: a complex picture emerges, Hamill [/bib_ref] [bib_ref] The short arm of the laminin gamma2 chain plays a pivotal role..., Gagnoux-Palacios [/bib_ref] [bib_ref] Globular domains 4/5 of the laminin alpha3 chain mediate deposition of precursor..., Sigle [/bib_ref] [bib_ref] Laminin 5 deposition regulates keratinocyte polarization and persistent migration, Frank [/bib_ref] [bib_ref] The role of alpha3beta1 integrin in determining the supramolecular organization of laminin-5..., Dehart [/bib_ref] [bib_ref] Integrin beta4 regulates migratory behavior of keratinocytes by determining laminin-332 organization, Sehgal [/bib_ref] [bib_ref] The beta3 chain short arm of laminin-332 (laminin-5) induces matrix assembly and..., Nakashima [/bib_ref]. These studies have shown that many factors including cell surface proteins, ECM proteins and intracellular signaling molecules are involved in the deposition and/or organization of laminin matrix. In this study, we could not detect any of type IV and VII collagens, perlecan and nidogen-1 in Lm332-ECM. Although the exact mechanism of laminin deposition remains to be clarified, it seems clear that secreted laminins can be deposited without support of any other ECM molecules. BM proteins such as nidogens, perlcan and type VII collagen are thought to stabilize the laminin matrix in vivo [bib_ref] Laminin deposition in the extracellular matrix: a complex picture emerges, Hamill [/bib_ref]. Cell surface receptors such as integrins, dystroglycans and sulfatides have been reported to regulate the organization and/or deposition of the laminin matrix [bib_ref] The role of alpha3beta1 integrin in determining the supramolecular organization of laminin-5..., Dehart [/bib_ref] [bib_ref] Integrin beta4 regulates migratory behavior of keratinocytes by determining laminin-332 organization, Sehgal [/bib_ref] [bib_ref] Laminin-sulfatide binding initiates basement membrane assembly and enables receptor signaling in Schwann..., Li [/bib_ref]. In the present study, the Lm332 deposition by migrating cells was independent of Lm332-binding integrins such as integrins a3ß1, a6ß1, and a6ß4, but stationary or confluent cells seemed to interact with the selfmade Lm332 matrix through these integrins, modulating the pattern of Lm332 matrix. On the other hand, sodium selenate, an effective inhibitor for the sulfation of heparan sulfates and chondroitin sulfates [bib_ref] Inhibition of synthesis of heparan sulfate by selenate: possible dependence on sulfation..., Dietrich [/bib_ref] , significantly inhibited the Lm332 deposition, suggesting that heparan sulfate proteoglycans such as syndecans might play an important role in this process. It seems also possible that sulfated glycolipids on cell membrane mediate the Lm332 deposition.
Full-sized laminins such as laminin-111, laminin-211 and laminin-511 are able to self-or co-polymerize in the matrix [bib_ref] Self-assembly of laminin isoforms, Cheng [/bib_ref] [bib_ref] The laminin alpha2 expressed by dystrophic dy(2J) mice is defective in its..., Colognato [/bib_ref]. Because the LN domains of the three full-sized laminin chains are critical for the polymerization, Lm332, of which the three chains are all truncated in their short arms, has been believed to be incapable of self-polymerization or co-polymerization with other laminins [bib_ref] Laminin deposition in the extracellular matrix: a complex picture emerges, Hamill [/bib_ref]. In the present study, Lm332-HEK cells deposited Lm332 in a mesh-like network structure as analyzed by electron microscopy. Although ß3c2-HEK cells deposited the ß3 and c2 proteins, probably in a heterodimer form, such a mesh structure was not found in the ß3c2-ECM. In addition, the deposited Lm332 matrix was not dissociated into the Lm332 heterotrimer by SDS in the absence of reducing reagent. These results strongly suggest that the Lm332 heterotrimer is able to self-polymerize in the matrix. It has been reported that the short arm of the c2 chain [bib_ref] The short arm of the laminin gamma2 chain plays a pivotal role..., Gagnoux-Palacios [/bib_ref] and the LG4-5 domain of the a3 chain [bib_ref] Globular domains 4/5 of the laminin alpha3 chain mediate deposition of precursor..., Sigle [/bib_ref] are important for the Lm332 deposition. By using a HEK cell line expressing Lm332 without the c2 short arm, we have confirmed that the short arm is critical for the Lm332 deposition (unpublished data). We also found that the LG4-5 domain of the a3 chain enhances the Lm332 deposition [bib_ref] Regulation of biological activity and matrix assembly of laminin-5 by COOH-terminal, LG4-5..., Tsubota [/bib_ref] but it seems not essential [bib_ref] Characterization of laminin 5B and NH 2 -terminal proteolytic fragment of its..., Kariya [/bib_ref]. The short arm of the ß3 chain does not significantly affect the Lm332 deposition but it promotes the deposition of laminin-511, suggesting its interaction with the full-length laminin chains [bib_ref] The beta3 chain short arm of laminin-332 (laminin-5) induces matrix assembly and..., Nakashima [/bib_ref]. Present analyses showed that after the deposition of Lm332, the c2 short arm and the a3 LG4-5 domain were released by proteolytic cleavage. This suggests that they are required for the Lm332 deposition but not directly involved in the polymerization. It is expected that the interaction of the c2 short arm with heparan sulfate proteoglycans or sulfated glycolipids plays a critical role in the efficient deposition of Lm332. Our results that Lm511 and Lm311 were not deposited on the matrix also imply the important role of the laminin c2 short arm in the Lm332 deposition. Further studies are required to clarify how the cleaved Lm332 selfpolymerizes in the matrix.
Although there were numerous studies on the biological activity of Lm332 protein, few reports compared the activities of Lm332-ECM and purified Lm332. Here we demonstrated that Lm332-ECM was clearly different from the purified, coated Lm332 substrate concerning the biological activity. It has been accepted that purified Lm332 efficiently supports cell adhesion and migration [bib_ref] Laminin-5 (laminin-332): Unique biological activity and role in tumor growth and invasion, Miyazaki [/bib_ref]. However, Lm332-ECM supported the adhesion of keratinocytes much more rapidly and strongly than the purified Lm332. The strong and stable cell adhesion to Lm332-ECM, which was evident by the resistance of keratinocytes to cell detachment treatments, lead to the suppressed cell migration. Both Lm332-ECM and ß3c2-ECM hardly supported cell migration, but treatment of these ECMs with purified Lm332 enhanced cell migration. Even in this experiment, the Lm332-ECM coated with purified Lm332 showed suppressed cell motility activity as compared with purified Lm332 alone or ß3c2-ECM plus Lm332. These results clearly indicate that Lm332-ECM rather inhibits cell migration.
The strong cell adhesion to Lm332-ECM obviously depends on the interaction with integrins. It has been reported that keratinocytes interact with self-deposited Lm332 through integrins a3ß1 for polarization and migration [bib_ref] Laminin 5 deposition regulates keratinocyte polarization and persistent migration, Frank [/bib_ref]. Integrin a3ß1 also functions for the proper organization of deposited Lm332 [bib_ref] The role of alpha3beta1 integrin in determining the supramolecular organization of laminin-5..., Dehart [/bib_ref]. Our results indicated that integrin a3ß1 bound to Lm332-ECM in a much higher affinity than purified Lm332. In addition, experiments with neutral integrin antibodies suggested that integrin a6ß4 contributed to the cell adhesion to Lm332-ECM more greatly than that to purified Lm332. The interaction of Lm332 with the cell surface integrin a6ß4 nucleates the stable cell adhesion structure hemidesmosome in keratinocytes [bib_ref] Biology and function of hemidesmosomes, Nievers [/bib_ref]. The present immunocytochemical staining with an anti-integrin ß4 antibody revealed numerous detergent-resistant, hemidesmosomelike structures in the cells adhered to Lm332-ECM, whereas these structures were almost absent in the cells adhered to purified Lm332. Based on all these data, it may be concluded that the polymerized Lm332 matrix strongly binds to integrins a3ß1 and a6ß4, and the latter binding promotes hemidesmosome formation, resulting in the tight and stable cell adhesion and the suppressed cell migration. Our finding is similar to the previous study that three-dimensionally organized fibronectin matrix has stronger cell adhesion activity than purified fibronectin [bib_ref] Taking cell-matrix adhesions to the third dimension, Cukierman [/bib_ref].
The stable adhesion of keratinocytes to Lm332-ECM mimics the interaction of Lm332 with integrin a6ß4 in the hemidesmosome structure of basal keratinocytes in vivo [bib_ref] Biology and function of hemidesmosomes, Nievers [/bib_ref]. Lm332 is known to bind to type VII collagen via the short arm of the ß3 chain, forming the anchoring filaments [bib_ref] Regulation of cell adhesion and type VII collagen binding by the beta3..., Nakashima [/bib_ref]. It is highly expected that the self-polymerization of Lm332 occurs in normal basement membranes and type VII collagen binds to the polymerized Lm332. These Lm332-dependent cell structures could support the stable anchoring of the edpidermis to the dermis in the normal skin. On the other hand, the cell motility activity of Lm332 is thought to contribute to wound repair [bib_ref] Cloning of the LamA3 gene encoding the alpha 3 chain of the..., Ryan [/bib_ref] and tumor invasion [bib_ref] Roles of laminin-332 and alpha6beta4 integrin in tumor progression, Kariya [/bib_ref]. We have previously reported that Lm332 stimulates cell migration in a soluble form [bib_ref] The basement membrane protein laminin-5 acts as a soluble cell motility factor, Kariya [/bib_ref]. It has been reported that matrix metalloproteinases capable of cleaving the short arm of the laminin c2 chain are overexpressed in pathological conditions such as wound healing and cancer invasion [bib_ref] Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5, Giannelli [/bib_ref] [bib_ref] Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5, Koshikawa [/bib_ref]. It is expected that the proteolytic cleavage of the c2 chain prevents the polymerization and assembly of Lm332 into the basement membrane. The resultant soluble Lm332, like coated Lm332, is likely to promote the migration of normal skin cells and cancer cells in the wound healing and cancer invasion, respectively [bib_ref] Laminin-5 (laminin-332): Unique biological activity and role in tumor growth and invasion, Miyazaki [/bib_ref].
In conclusion, the present study strongly suggests that the contrasting activities of Lm332, i.e. the stable cell adhesion vs. the enhanced cell motility, are caused from the different states of Lm332, i.e. polymerized Lm332 matrix and nonpolymerized soluble Lm332. It is assumed that the soluble or unassembled form of Lm332 plays an important role in the elevated cell migration during the wound healing and tumor invasion.
# Materials and methods
## Antibodies and reagents
Mouse monoclonal antibodies against the N-terminal regions of human laminin a3A chain (LSac3 and BG5), ß3 chain (12C) and c2 chain (D4B5) and one against human nidogen-1 (10F) were produced in our laboratory [bib_ref] Localization of laminin alpha3B chain in vascular and epithelial basement membranes of..., Kariya [/bib_ref]. Function-blocking anti-integrin antibodies used are anti-a2 integrin antibody (P1E6), anti-a3 integrin antibody (P1B5), anti-a5 integrin antibody (P1D6) and anti-ß1 integrin antibody (6S6) from Chemicon (Temecula, CA), and anti-a6 integrin antibody (GoH3) from PharMingen (San Diego, CA). For immunostaining, anti-laminin-c2 chain antibody (GB3) and anti-a6 integrin (H-87) and anti-ß4 integrin (H-101) antibodies from Santa Cruz Biotechnology (Santa Cruz, CA) were also used. Other commercial antibodies to human antigens used for immunoblotting are monoclonal antibody against the laminin ß3 chain (Kalinin B1) from Transduction Laboratories (Lexington, KY), rabbit polyclonal antibody against the laminin a5 chain (LAMA A01) from Abnova (Taipei, Taiwan), anti-fibronectin antibodies [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] from Takara (Tokyo, Japan), anti- . Differential detachment of NHK cells adhered to purified Lm332 and Lm332-ECM. NHK cells were inoculated onto the plates coated with 1 mg/ml purified Lm332 or deposited with Lm332-ECM and incubated for 1 h. After washing with PBS, the cells were incubated with trypsin/EDTA diluted 1:35 in PBS (A) or with 10 mM EDTA alone (B). After incubation for the indicated lengths of time and then washing with PBS, the relative number of adherent cells was determined as described in [fig_ref] Figure 5: Fine structures of Lm332-ECM [/fig_ref]. Each bar indicates the mean 6 S.D. of the fluorescent intensity (FI) for adherent cells in triplicate assays. doi:10.1371/journal.pone.0035546.g009 type VII collagen antibody (LH7.2) from Sigma (St. Louis, MO), anti-type IV collagen antibody (H-234) from Santa Cruz, and antiperlecan antibody (Clone 7B5) and Alexa Fluor 488-labelled secondary antibody from Invitrogen (Camarillo, CA). Human recombinant Lm332 and Lm311 were purified as described previously [bib_ref] Efficient expression system of human recombinant laminin-5, Kariya [/bib_ref] [bib_ref] Laminin-3B11, a novel vascular-type laminin capable of inducing prominent lamellipodial protrusions in..., Mori [/bib_ref]. Human recombinant Lm511 was purchased from BioLamina (Sundbyberg, Sweden).
## Cells and transfectants
Human embryonic kidney cell line HEK293 (ATCC CRL-1573) was purchased from American Type Culture Collection and transfected with the cDNAs of the three Lm332 subunits to overexpress the wild-type Lm332 (Lm332-HEK) [bib_ref] Efficient expression system of human recombinant laminin-5, Kariya [/bib_ref] or an a3mutated Lm332 resistant to the proteolytic processing of the a3 chain (Lm332a3AA-HEK) [bib_ref] Regulation of biological activity and matrix assembly of laminin-5 by COOH-terminal, LG4-5..., Tsubota [/bib_ref]. Lm332-producing human cancer cell lines used were epidermoid carcinoma of the vulva (A431), epidermoid carcinoma of the cervix (CaSki), squamous adenocarcinoma of the tongue (HSC-4) and gastric adenocarcinomas (STKM-1 and MKN-45). STKM-1 was established and provided by Dr. S. Yanoma (Kanagawa Cancer Center, Yokohama, Japan) [bib_ref] Establishment and characterization of a CA19-9 producing human gastric cancer cell line,..., Arimura [/bib_ref] , and the others were obtained from Japanese Cancer Resources Bank (JCRB; Tokyo). Expression of Lm332 in these cancer cell lines were reported in our past studies [bib_ref] A large cell-adhesive scatter factor secreted by human gastric carcinoma cells, Miyazaki [/bib_ref] [bib_ref] Differential expression of laminin-5/ladsin subunits in human tissues and cancer cell lines..., Mizushima [/bib_ref]. All these cell lines were stored in a liquid N 2 tank in our laboratory and cultured in DMEM/F12 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal calf serum, penicillin and streptomycin sulfate. NHK cells from neonatal foreskin were obtained from Cascade Biologics (Portland, OR), and cultured in KGM medium (Sanko-Junyaku, Tokyo, Japan), which was composed of keratinocyte basal medium, 0.1 ng/ml human EGF, 0.4% (v/v) bovine pituitary extract, 10 mg/ml insulin, 500 ng/ml hydrocortisone, 50 mg/ml gentamicin, and 50 ng/ml amphotericin B. Passages 2 and 3 were used in the described experiments.
## Preparation of cm, deposited ecm, and lm332-coated plates
Unless otherwise noted, cells were grown to subconfluence in the growth medium, washed three times with PBS and then incubated in serum-free medium. Two days later, the CM was collected and added with two protease inhibitors, phenylmethylsulfonyl fluoride and N-ethylmaleimide. The CM was dialyzed against pure water, lyophilized and then dissolved in a 1/50 volume of PBS. To prepare deposited ECM, subconfluent cultures were incubated in the growth medium for 2 days with medium change every day. After washing twice with PBS, the cells were removed from the plates by incubating with 10 mM EDTA, and the plates were washed five times with PBS and then used for the following assays. In cases of NHK cells, the cells remaining on the plates after EDTA treatment were completely removed by further treating with 20 mM NH 4 OH for 5 min. All preparations were checked to be cell-free under a microscope. For immunoblotting analysis, the deposited ECM was dissolved in the SDS sample buffer. In some experiments, cells were directly inoculated and incubated at a high density in serum-free medium for the indicated lengths of time, and ECM and/or CM were prepared. To coat culture plates with purified Lm332 protein or CM, the plates were incubated with Lm332 or CM overnight at 4uC, briefly washed with PBS, and blocked with 1% bovine serum albumin (BSA) at room temperature for 1 h. After washing three times with PBS, the plates were used for the following assays.
ELISA ELISA was carried out as follows. Ninety-six-well plates coated or deposited with test substances were blocked with 2% BSA for 1 h, washed three times with PBS containing 0.1% Tween20 (PBS/Tween), and then incubated with anti-a3 (Lsac3) or anti-c2 chain (D4B5) antibody (diluted 1:1000 with PBS/Tween) for 1 h at room temperature. After washing with PBS/Tween, the samples were incubated with goat anti-mouse IgG antibody coupled with biotin and then with alkaline phosphatase-conjugated avidin D. The immunosignals were visualized with pnitrophenylphosphate and measured for absorbance at 405 nm.
## Immunofluorescence staining of lm332 and integrins
To analyze the Lm332 deposition by cultured cells, Lab-Tek 8well chamber slides (Nunc, Naperville, IL) were previously coated with 10 mg/ml bovine type I collagen (Koken, Tokyo, Japan) at 4uC overnight and washed with PBS. Cell suspension (2610 3 cells/0.25 ml) in serum-free medium was inoculated per well of the chamber slides and incubated for 6 h. The cultures were washed with PBS, fixed with 4% (w/v) paraformaldehyde in PBS for 10 min, and then treated with 0.5% (v/v) Triton X-100 in PBS for 15 min. The fixed cells were blocked with 2% BSA in PBS for 1 h and then reacted with the mouse anti-laminin a3 chain antibody BG5 or other antibodies for 1 h. After washing with PBS, the cultures were incubated with a mixture of FITC-conjugated goat anti-mouse IgG antibody (Vector Laboratory, Burlingham, CA) and rhodamine phalloidin (Invitrogen) for 1 h, and then washed with PBS. Fluorescence images were obtained using fluorescence microscope BZ-8000 (Keyence, Osaka, Japan) or a LSM510 confocal microscope (CarlZeiss). To immunostain Lm332 in the ECMs deposited by various types of cells, the ECMs were prepared as described above and directly subjected to the staining without the fixation. In the analysis of integrin localization, rabbit polyclonal antibodies against ß4 integrin (H-101) and a6 integrin (H-87) were used as primary antibodies.
## Cell adhesion assay
Cell adhesion assay with NHK cells was performed as described previously [bib_ref] Differential regulation of cellular adhesion and migration by recombinant laminin-5 forms with..., Kariya [/bib_ref]. Briefly, each well of 96-well ELISA plates (Costar, Cambridge, MA) was coated with a substrate protein at indicated concentrations at 4uC overnight and then blocked with 1% BSA. Cells (2610 4 cells) were inoculated per well containing KGM medium, and incubated in described conditions. After nonadherent cells were removed, adherent cells were fixed and stained with Hoechst 333432. The fluorescent intensity of each well of the plates was measured using a CytoFluor 2350 fluorometer (Millipore, Bedford, MA). For inhibition assay, the cell suspension was incubated with function-blocking anti-integrin antibodies or inhibitors under the indicated conditions before inoculation.
## Cell migration assay
NHK cells (2.5610 4 cells in KGM medium) were inoculated per well of 24-well plates pre-coated with a test protein or deposited with cell-derived ECM. After pre-incubation for 1.5 h at 37uC, cell movement was monitored using a time-lapse video equipment for 5.5 h. Total length of random pass that each cell covered was measured using a video micrometer (VM-30, Olympus, Tokyo).
## Sds-page and immunoblotting
SDS-PAGE was performed on 5% gels, or 4.0-7.5% or 5.0-20% gradient gels under reducing or non-reducing conditions. Separated proteins were stained with CBB. For immunoblotting analyses, proteins resolved by SDS-PAGE were transferred to nitrocellulose membranes and detected with the ECL detection reagents (GE Healthcare, Buckinghamshire, UK).
## Integrin binding assay
Integrin titration assays were carried out by the method of Nishiuchi et al. [bib_ref] Characterization of the ligand-binding specificities of integrin alpha3beta1 and alpha6beta1 using a..., Nishiuchi [/bib_ref]. Microtiter plates were coated with 1 mg/ml Lm332 overnight at 4uC or deposited with Lm332-ECM as described above. The wells were blocked with 1.2% BSA at room temperature for 1 h and then washed with TBS/Mn (20 mM Tris-HCl, pH7.5, 0.15 M NaCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 1 mM MnCl 2 ) containing 0.1% BSA and 0.02% Tween-20 (Buffer A). Serially diluted a3ß1 integrin with Buffer A was added to the plates and allowed to bind to the substrates for 3 h. For negative control, Buffer A containing 10 mM EDTA was used. The plates were washed with 25 mM HEPES (pH 7.6) containing 1 mM MnCl 2 or 10 mM EDTA, and bound integrins were fixed with 2.5% glutaraldehyde in HEPES buffer for 10 min. The plates were washed with TBS/Mn, and the bound integrin was quantified by ELISA. Buffer A was used for the dilution of reagents and plate washing. The absorbance obtained in the presence of 10 mM EDTA was subtracted as background from each data.
## Cell detachment assay
NHK cells (2610 4 cells) were seeded into each well deposited with Lm332-ECM or coated with purified Lm332 in 96-well plates, and incubated for 1 h at 37uC. The cells were then treated with a solution of trypsin/EDTA (Cambrex Bio Science, Walkersville) diluted 1:35 in PBS or with 10 mM EDTA/PBS for varied lengths of time. The relative number of adherent cells was determined as described in the cell adhesion assay section.
## Transmission electron microscopy of ecm proteins
Confluent cultures of Lm332-HEK and ß3c2-HEK cells were incubated on poly-L-lysine-coated cover slide glasses for 4 days, and their deposited ECMs were prepared as described above. The ECMs were then fixed with 2% glutaraldehyde and then with osmium tetraoxide. The materials were analyzed with JEOL JEM 200EX (Tokyo) at Hanaichi Electron Microscope Technology Institute (Okazaki, Japan). [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] Immunostained patterns of Lm332 matrices deposited by normal keratinocytes (NHK) and four cancer cell lines (A431, HSC-4, STKM-1 and MKN-45). Each kind of cells (1610 5 cells) were inoculated per well of Lab-Tek 8-well chamber slides in serum-free medium and incubated for 2 days. After the cells were removed by treating with 10 mM EDTA, deposited Lm332 matrices were immunostained with the anti-laminin a3 chain antibody BG5 and a FITC-conjugated secondary antibody. Other experimental conditions are described in ''Materials and Methods''. Bars, 100 mm. (TIF)Immunostaining of Lm332 matrices deposited by NHK and Lm332-HEK cells with antibodies to the laminin a3 (BG5), ß3 (12C) and c2 (GB3) chains. The Lm332 matrices deposited by the two types of cells during 6 h incubation were subjected to immunofluorescence staining with the three different antibodies. Other experimental conditions are the same as described in [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref]. (TIF) Effect of anti-integrin antibodies on Lm332 deposition by Lm332-HEK cells. (A) Effect on cell attachment. Ninety-six-well plates were coated with 0.3 mg/ml purified Lm332 and blocked with BSA. Lm332-HEK cells suspended in serum-free medium were pretreated with non-immune mouse IgG (20 mg/ml) as a negative control or with both anti-a3 integrin (P1B5) and anti-a6 integrin (GoH3) antibodies (20 mg/ml IgG each) at 37uC for 15 min. The pretreated cells were inoculated onto the Lm332-coated plates and incubated for 1 h. After the incubation, adherent cells were determined. Each bar represents the mean 6 S.D. of the fluorescent intensity (FI) for adherent cells in triplicate assays. (B) Effect on Lm332 deposition. Lm332-HEK cells treated with the control IgG (left panel) or with the antiintegrin antibodies (right panel) were inoculated on collagencoated 8-well chamber slides and incubated for 6 h. The cultures were then stained for Lm332 with the anti-a3 chain antibody BG5 followed by a FITC-labeled secondary antibody (green) and for Factin with rhodamine phalloidin (red). Other experimental conditions are described inand ''Materials and Methods''. (TIF) [fig_ref] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells [/fig_ref] Quantitative assay of Lm332 deposited on culture plates by Lm332-HEK cells by ELISA and CBB staining. Fifty ml of purified Lm332 protein (open circles) were coated at the indicated concentrations to the 96-well plates. Lm332-HEK transfectant was cultured in DMEM/F12 medium supplemented with 10% fetal calf serum, and ECM proteins (closed circles) were deposited on the plates for 3 days. The amount of Lm332 on the plates was determined by ELISA using the antibodies against the laminin a3 (A) and c2 (B) chains. Each bar represents the mean 6 S.D. for triplicate assays. The data shown are representative of at least three independent experiments performed. The Lm332 concentration on the plate was equivalent to that obtained by coating purified Lm332 at a concentration of 0.61 mg/ml or 0.56 mg/ml as analyzed for the a3 and c2 chain, respectively. (C) A 90-mm culture dish was coated with 10 ml of 1.0 mg/ml Lm332, while another 90-mm dish was deposited with Lm332-ECM by Lm332-HEK cells as described above. The coated Lm332 and the deposited Lm332-ECM were collected by dissolving with the SDS sample buffer. A 1/3 aliquot of each extract was run on a 5-20% gradient gel and stained with CBB. The ratio of the total band intensity of Lm332-ECM to the purified Lm332 was determined to be 1.1 by the NIH image software. (TIF) [fig_ref] Figure 5: Fine structures of Lm332-ECM [/fig_ref] Immunoblotting analyses of Lm511, nidogen-1 and fibronectin present in CM and ECM of Lm322-HEK cells. CM (lane 1) and ECM (lane 2) were prepared from the confluent culture of Lm332-HEK cells incubated for 3 days in serum-free medium and subjected to immunoblotting, as described in [fig_ref] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes [/fig_ref] and ''Materials and Methods''. In both cases, approximately 5% of the total sample was applied to each lane of SDS-PAGE. (A) Immunoblots for Lm511 subunits. The CM and ECM were analyzed for the laminin a3, a5, ß1 and c1 chains.
## Supporting information
[fig] Figure 1: Deposition of Lm332 on culture plates by normal keratinocytes (NHK) and five cancer cell lines (A431, CaSki, HSC-4, STKM-1 and MKN-45) [/fig]
[fig] Figure 2: Immunofluorescent staining of Lm332 deposited by three cell lines. NHK (A, top), HSC-4 cells (A, center) and Lm332-HEK cells (A. bottom) were suspended in serum-free medium, inoculated at a cell density of 2610 3 cells/well on collagen-coated 8-well chamber slides and incubated for 6 h. The cultures were stained for F-actin with rhodamine phalloidin (left panels) and for Lm332 with the anti-a3 chain BG5 antibody and followed by a FITC-labeled secondary antibody (center panels), as described in ''Materials and Methods''. Right panels are merged images. In (B), Lm332-HEK cells were inoculated at a high cell density (1610 5 cells/well), incubated for 6 h (left panel) or 48 h (right panel), and stained for Lm332 as above. doi:10.1371/journal.pone.0035546.g002 [/fig]
[fig] Figure 4: SDS-PAGE analyses of Lm332-ECM deposited by Lm332-HEK cells. Confluent cultures of Lm332-HEK (lane 2), a3AA-Lm332-HEK (lane 3) and ß3c2-HEK (lane 4) were incubated in the serum-containing growth medium for 4 days, and the resultant ECMs were prepared and applied to SDS-PAGE under reducing conditions, followed by CBB staining (A) or immunoblotting with anti-a3 (B, upper panel) and -c2 (B, lower panel) chain antibodies. Purified Lm332 was run as a standard on lane 1. Bars on the left indicate major protein bands with their approximate molecular sizes in kDa. The two minor bands (# and *) in (A) were identified as laminin c2 fragments by NH 2 -terminal amino acid sequencing. Other experimental conditions are described in ''Materials and Methods''. (C) To see the proteolytic processing of deposited Lm332, Lm332-HEK cells were incubated for 6 h and 30 h, and the deposited Lm332 was prepared and applied to SDS-PAGE under reducing conditions, followed by immunoblotting for the a3 chain (left panels) and the c2 chain (right panels). (D) CM (lane 1) and ECM (lane 2) were prepared from the confluent culture of Lm332-HEK cells after incubation in serum-free medium for 2 days and analyzed by immunoblotting under non-reducing conditions with anti-a3 (left panel) and -ß3 (right panel) chain antibodies. Lane 3, purified Lm332 with processed c2 chain (400 kDa). Note that Lm332 in the ECM remains on the gel top as a broad band (lane 2). doi:10.1371/journal.pone.0035546.g004 [/fig]
[fig] Figure 5: Fine structures of Lm332-ECM (left), ß3c2-ECM (center) and purified Lm332 (right) were analyzed by TEM at a magnification of 6 50,000. Bars indicate 100 nm. doi:10.1371/journal.pone.0035546.g005 [/fig]
[fig] Figure 6: Migration and morphology of NHK cells on purified Lm332 and Lm332-ECM substrates. (A) NHK cells were inoculated on [/fig]
[fig] Figure 7: Cell adhesion activity of Lm332-ECM and purified Lm332 toward NHK cells. ECMs from ß3c2-HEK and Lm332-HEK cultures and Lm332-coated plates were prepared as described inFigure 4. (A) Phase-contrast microscopic images of NHK cells after 10 min incubation. Original magnification, 6300. Lm332 protein was coated at 1.0 mg/ml on the plate. (B) NHK cells suspended in KGM growth medium were inoculated into each well and then incubated at 37uC for 10 min. After the incubation, non-adherent cells were removed, and adherent cells were quantified. Numerical values under three right columns indicate the concentration at mg/ml of coated Lm332 protein. Each bar represents the mean 6 S.D. of the fluorescent intensity (FI) for adherent cells in triplicate assays. The data shown are representative of at least three independent experiments performed. doi:10.1371/journal.pone.0035546.g007 [/fig]
[fig] Figure 8: Interaction of purified Lm332 or Lm332-ECM with integrins. (A and B) To identify integrins responsible for cell adhesion, NHK cells suspended in the KGM medium were pretreated with the indicated integrin antibodies (2 mg/ml IgG) or EDTA for 30 min at room temperature and plated on the plates with purified Lm332 (A) or Lm332-ECM (B). After 20 min incubation, adherent cells were quantified. The relative number of adherent cells in the presence of control mouse or rat IgG was taken as 100%. Each bar represents the mean 6 S.D. of the fluorescent intensity (FI) for adherent cells in triplicate assays. (C) Morphology of NHK cells was examined after incubation on Lm332-ECM in the presence of control mouse IgG, or the indicated anti-integrin neutral antibodies for 20 min. In the upper left culture (Lm332+IgG), NHK cells were incubated in a well precoated with 1.0 mg/ml purified Lm332 in the presence of the control IgG. Original magnification, 6300. The images shown are representative of at least three independent experiments performed. (D) Binding affinity of integrin a3ß1 to purified Lm332 and Lm332-ECM was determined by ELISA. Varied concentrations of purified integrin a3ß1 was allowed to bind to the plates coated with 1 mg/ml purified Lm332 (closed squares) or deposited with Lm332-ECM (open circles) in the presence of 1 mM MnCl 2 . Bound integrin a3ß1 was quantified by ELISA using an anti-integrin a3ß1 polyclonal antibody. The amounts of integrin a3ß1 bound in the presence of 10 mM EDTA was taken as nonspecific binding and subtracted as the background. The results shown are the means of duplicate assays. doi:10.1371/journal.pone.0035546.g008 [/fig]
|
Associations Between Cancer Screening Behavior and Complementary Medicine Use: Results of a National Cross-Sectional Survey of 9151 Australian Women
Introduction: Complementary medicine (CM) use has been found to influence the uptake of conventional cancer treatment. This study examines associations between CM use and cancer screening rates. Methods: Women aged 62 to 67 years from the Australian Longitudinal Study on Women's Health were surveyed regarding their use of cancer screening initiatives. Associations between cancer screening behavior and visits to CM practitioners were analyzed. Results: Of the 9151 women, 9049 (98.9%) completed questions about cancer screening. A total of 65.1% of women had received a clinical skin examination, 54.3% colorectal cancer screening, 56.2% Pap test (within past 2 years), 83.3% mammogram (within past 2 years), 55.8% clinical breast examination, and 55.8% had conducted breast self-examination. Women who had consulted a massage therapist were more likely to undergo clinical skin examination (P = .002), clinical breast examination (P = .018), and mammogram (P = .001). Women who had consulted a chiropractor were more likely to undergo a clinical skin examination (P = .001), colorectal cancer screening (P = .020), and mammogram (P = .011). Women who had consulted an acupuncturist were more likely to undergo colorectal cancer screening (P = .019), and those who consulted with an osteopath were more liable to have a Pap test (P = .049). Conclusion: Women who visit CM practitioners are more likely to participate in cancer screening initiatives. Research is required to understand the current and potential role that CM practitioners (can) have as public health advocates, recommending preventative health measures such as cancer screening. Such an examination will help ensure optimal screening utilization and effective, timely care for all cancer patients.
# Introduction
Cancer screening is an important public health initiative that enables early detection of cancer or precancerous changes. In Australia, the current Guidelines for Preventative Activities in General Practice (ninth edition) published by the Royal Australian College of General Practitioners (2016) recommends performing a Papanicolaou (Pap) test every 2 years.The screening recommendation for Australian's National Cervical Cancer Screening Program will change in December 2017 to recommend women aged 25 to 74 years have a Pap test every 5 years.A mammogram is recommended every 2 years for low-risk women, and colorectal cancer screening with occult blood testing at least every 2 years, for women aged 50 to 64 years. General practitioners (GPs) are also recommended to regularly enquire with patients about skin changes and early signs of skin cancer 1 due to the high rate of melanoma in Australia. [bib_ref] The growing burden of invasive melanoma: projections of incidence rates and numbers..., Whiteman [/bib_ref] Despite the availability of cancer screening to detect common forms of cancer and strong general practice guidelines, many women do not partake in these screening programs. In 2016, it was predicted that 13 280 new cases of melanoma would be diagnosed in Australia, of which 41% will be in women.However, the number of women participating in skin cancer screening in Australia is not recorded.Breast cancer is the most common cancer affecting Australian women, and BreastScreen Australia is the national breast cancer screening program that primarily targets women aged between 50 and 74 years.In 2014-2015, 54% of Australian women in the target range participated in breast screening (screening conducted in BreastScreen Centers by breast screening doctors and nurses throughout Australia).While the rates of cervical cancer are much lower than breast and skin cancer in Australia (around 7 in every 100 000 women), it is largely a preventable disease. However, only 57% of women aged 20 to 69 years participate in Pap testing.The National Bowel Cancer Screening Program provides free colorectal cancer screening for adults in targeted age groups, and 47% of women aged 60 to 64 years participated in 2013-2014.The use of complementary medicine (CM)-a diverse group of products and practices located outside of the dominant medical system of training and practice 7 -has increased in recent years. [bib_ref] Gender, symptom experience, and use of complementary and alternative medicine practices among..., Fouladbakhsh [/bib_ref] CM users are more frequently female and older (mid-age) and are higher users of health care services generally when compared with adults who do not use CM. [bib_ref] Gender, symptom experience, and use of complementary and alternative medicine practices among..., Fouladbakhsh [/bib_ref] While research demonstrates that women with cancer are high users of CM, [bib_ref] Gender, symptom experience, and use of complementary and alternative medicine practices among..., Fouladbakhsh [/bib_ref] [bib_ref] Naturopathy/herbalism consultations by mid-aged Australian women who have cancer, Adams [/bib_ref] [bib_ref] Who uses? A narrative review of demographic characteristics and health factors associated..., Bishop [/bib_ref] only limited research to date has contemplated the association between participation in cancer screening initiatives and the use of CM. A review of CM use and early breast cancer detection found 2 studies that reported a positive relationship between CM use and mammography, while another 2 studies found no relationship. [bib_ref] The relationship between complementary and alternative medicine use and breast cancer early..., Dale [/bib_ref] Additionally, one study found that women were less likely to have a mammogram if they saw a naturopath and more likely if they used massage therapy. [bib_ref] The relationship between complementary and alternative medicine use and breast cancer early..., Dale [/bib_ref] While GP visits have been found to be predictive of increased health checks including cancer screening, [bib_ref] Adherence to recommended health checks by women in mid-life: data from a..., Byles [/bib_ref] we need to further understand the relationship between women's consultations with CM practitioners and participation in broader cancer screening. This study was conducted to advance our understanding of the relationship between CM use and cancer screening, in Australian women aged 62 to 67 years.
# Methods
# Survey methodology
The study analyzed data from the Australian Longitudinal Study on Women's Health (ALSWH), which has been designed to assess the health, well-being, and associated factors of Australian women over a 20-year period. Women in 3 different age groups (18-23, 45-50, 70-75 years) were randomly selected from the national Medicare database in 1996. [bib_ref] Women's Health Australia: recruitment for a national longitudinal cohort study, Brown [/bib_ref] Respondents have been shown to be broadly representative of the national population of women in their respective age cohorts. 14 Questionnaires were mailed to survey participants every 3 years on average. For this substudy on CM use and cancer screening, analyses focused on 9151 women from the mid-age ALSWH cohort, who were aged between 62 and 67 years at the time of the 2013 survey.
## Cancer screening rates and initiatives
The questionnaire included items on the following cancer screening initiatives: (a) clinical skin examination by a doctor (eg, spots, lesions, and moles) within the past 3 years; (b) colorectal cancer screening within the last 3 years; (c) Pap test (date of most recent test); (d) mammography (date of the last mammogram); (e) clinical breast examination (breast examined by a doctor or a nurse); or (f) regular monthly breast self-examination. Women could answer yes or no to questions a, b, e, and f; questions c and d were dichotomized into "received it within the past 2 years" or "did not receive it" as per official recommendation for Pap tests and mammograms.
## Complementary and alternative medicine utilization
The questionnaire included items on consultations with CM practitioners within the previous 12 months, including massage therapists, naturopaths/herbalists, chiropractors, acupuncturists, or "other" alternative health practitioners (eg, aromatherapists, homoeopaths, reflexologists, and iridologists). Women could answer yes or no to these questions.
## Ethics
The ALSWH has been approved by the University of Newcastle's Human Research Ethics Committee (H-076-0795 and H-2012-0256) and the University of Queensland's Medical Research Ethics Committee (2004000224 and 2012000950). All participants gave written informed consent.
## Statistical analyses
Chi-squared tests were used to examine the association between cancer screening and consultation with a CM practitioner. Multiple logistic regression analyses were conducted to determine whether CM practitioner visits (independent variables) were associated with having participated in cancer screenings (dependent variables). Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were computed for all predictor variables. Analyses were adjusted for sociodemographic characteristics and known confounding variables (marital status, education, income, the area of residence, consultation with family doctors/GPs, hospital doctors or specialists, being diagnosed with chronic diseases, mental health conditions, or cancer). Statistical significance was set at P < .05. All statistical analyses were performed using IBM SPSS software (IBM SPSS Statistics for Windows, release 22.0; IBM Corp, Armonk, NY).
# Results
Of the 9151 women who completed the questionnaire, 9049 provided data on cancer screening. Of these women, 65.1% had received a clinical skin examination and 54.3% had participated in colorectal cancer screening within the past 3 years. Within the past 2 years, 56.2% of women had undertaken a Pap test, 83.3% had a mammogram, and 55.8% had a clinical breast examination or had conducted breast self-examination. [fig_ref] Table 1: The Bivariate Association Between Consulting Complementary Medicine Practitioners and Cancer Screening Behaviors,... [/fig_ref] shows the associations between consulting a CM practitioner and cancer screening rates. Significant differences were found between women who did and did not visit a CM practitioner and participation in cancer screening. Women who consulted a naturopath/herbalist, massage therapist, chiropractor, osteopath, and acupuncturist were more likely to have a clinical skin examination (P < .05). Similarly, women who visited a naturopath/herbalist, massage therapist, osteopath, and acupuncturist were more likely to have a clinical breast examination (P < .05). Women who visited a chiropractor or an acupuncturist were more likely to participate in colorectal screening (P < .05).
Women who had consulted a massage therapist or chiropractor were more likely to have a mammogram (P < .05); conversely, women who consulted a naturopath/herbalist, acupuncturist, or "other" alternative health practitioner in the last 12 months were less likely to have had a mammogram (P < .05).
The outputs from the adjusted logistic regression models are presented in [fig_ref] Table 2: Output From the Logistic Regression Models Showing the Association Between Consulting Health... [/fig_ref]. Women who had consulted a massage therapist were more likely to undergo a clinical skin examination (OR = 1.23; 95% CI = 0.98, 1.55; P = .002), clinical breast examination (OR = 1.15; 95% CI = 1.02, 1.29; P = .018), and mammogram (OR = 1.32; 95% CI = 1.12, 1.55; P = .001). Women who had consulted a chiropractor in the previous 12 months were more likely to undergo a clinical skin examination (OR = 1.28; 95% CI = 1.10, 1.49; P = .001), colorectal cancer screening (OR = 1.18; 95% CI = 1.03, 1.35; P = .020), and mammogram (OR=1.30; 95% CI = 1.06, 1.59; P = .011). Additionally, women who had consulted an acupuncturist were more likely to undergo colorectal cancer screening (OR = 1.29; 95% CI = 1.04, 1.61; P = .019), and those who consulted an osteopath were more likely to have a Pap test (OR = 1.27; 95% CI = 1.00, 1.62; P = .049).
# Discussion
This study reports the relationship between CM practitioner consultation and participation in cancer screening in a large cohort of Australian women. Our analyses reveal several interesting findings. First, compared with women who did not consult a CM practitioner and women who consulted a CM practitioner had significantly higher rates of clinical skin examination and clinical breast examination.
Conversely, a US population study of more than 16 000 adults found the use of both conventional medical (defined as office-based GP services and outpatient hospital-based physician visits) and CM services negatively associated with clinical breast examination, [bib_ref] Association between use of unconventional therapies and conventional medical services, Druss [/bib_ref] as compared with women who used conventional services only. Similarly, an Australian study found no relationship between CM product use and clinical breast examination; however, visits to CM practitioners were not evaluated [bib_ref] Associations between the use of complementary and alternative medications and demographic, health..., Gollschewski [/bib_ref] and associations remain unknown. Further detailed examination of cancer screening behavior of CM users, along with the preferences and cancer screening recommendations of CM practitioners is warranted to help ensure increased uptake of effective cancer screening strategies among women.
Women who consult CM practitioners may engage in proactive health behavior more frequently, 7 possibly including timely cancer screening practices. Research describes CM users as being proactive consumers who often have a holistic approach to health that involves active participation, [bib_ref] Self-care dimensions of complementary and alternative medicine use among older adults, Votova [/bib_ref] and they may equally participate in cancer screening as a proactive health measure. It is also possible that a CM practitioner may recommend cancer screening in line with the preventative health philosophy of their discipline.Risk perception has been found to be predictive of both preventative health behavior [bib_ref] Does heightening risk appraisals change people's intentions and behavior? A meta-analysis of..., Sheeran [/bib_ref] and CM use. [bib_ref] Complementary and alternative medicine use is associated with an increased perception of..., Rakovitch [/bib_ref] A study involving breast cancer patients and their use and beliefs about CM found that women who used CM perceived the risk of cancer recurrence or death from cancer to be significantly greater than non-CM users. [bib_ref] Complementary and alternative medicine use is associated with an increased perception of..., Rakovitch [/bib_ref] As our study found an association between women who use CM and women who have clinical breast checks, these behaviors may both, at least in part, relate to concern about cancer occurrence or recurrence.
Increased health literacy may partly explain greater uptake of cancer screening initiatives by women who use CM in our study. CM users have frequently been found to have higher levels of education when compared with non-CM users. [bib_ref] Who uses? A narrative review of demographic characteristics and health factors associated..., Bishop [/bib_ref] [bib_ref] A prospective, multicenter study of complementary/alternative medicine (CAM) utilisation during definitive radiation..., Moran [/bib_ref] This finding suggests that women who consult CM practitioners may have higher levels of health literacy (the degree to which individuals can obtain, process, and understand basic health information to make appropriate health decisions). [bib_ref] Health literacy and cancer screening: a systematic review, Oldach [/bib_ref] Increased health literacy may extend to a better understanding of the benefits and availability of various cancer screening initiatives. In line with this, a recent systematic review evaluating the impact of health literacy on the uptake of cancer screening found an association between inadequate health literacy and lower cancer screening rates. [bib_ref] Health literacy and cancer screening: a systematic review, Oldach [/bib_ref] Further research is required to understand the role and impact of health literacy on women's knowledge and consequent decisions about cancer screening.
The results for mammography from our study were slightly more nuanced than those reported for other breast cancer screening techniques. Women who had consulted a massage therapist or chiropractor in the previous 12 months were more likely to have had a mammogram and women who had consulted a naturopath/herbalist, acupuncturist, or "other" alternative health practitioner in the same period were less likely to have received a mammogram. Similarly, a US study found a positive association between massage therapy and mammogram screening, while women who consulted naturopaths were less likely to have a mammogram. [bib_ref] Preventive screening of women who use complementary and alternative medicine providers, Downey [/bib_ref] Downey et al 23 also investigated differences between alternative medicine (defined as using CM instead of conventional biomedical care) and complementary therapy (defined as using CM alongside conventional biomedical care). They found women who used CM as a complement to conventional care were more likely to have a mammogram as opposed to women who had not visited a doctor. Many women who use CM may be high health care consumers generally, including conventional care. There may also be important distinctions between different types of CM users that could be further elucidated in future research.
Inherent differences between CM disciplines such as massage therapy and naturopathy may also partly explain the differences in mammography uptake. For example, women who consult practitioners in disciplines defined by philosophy, as naturopathy and acupuncture often are, may prefer a less intrusive and more natural method of breast examination in line with the philosophical tenants of the discipline. [bib_ref] Associations between the use of complementary and alternative medications and demographic, health..., Gollschewski [/bib_ref] Furthermore, research shows that many women consult CM practitioners due to a desire for treatment they consider "safe," "holistic," and "natural."The finding that women who seek care from naturopaths/herbalists and acupuncturists have a lower uptake of mammography may in some cases, be an extension of these beliefs and health seeking ideologies and a wish for a more noninvasive test.
Having confidence in nontraditional cancer treatments has previously been found to be a barrier to participation in mammography screening for minority women in the United States. [bib_ref] Barriers related to mammography use for breast cancer screening among minority women, Alexandraki [/bib_ref] However, more research is required to determine if this barrier exists for Australian women who have not had a mammogram and who consult a naturopath/herbalist or acupuncturist. Conversations between CM practitioners and their female patients may also influence women's decisionmaking around participation in mammogram screening, and further research is required to understand the daily routine care approaches of CM practitioners regarding recommending cancer screening.
Our study also highlights that women are more likely to have a clinical skin examination if they consult with either a massage therapist or a chiropractor. As the type of treatment provided by these practitioners involves touch and an opportunity to view a person's skin, it is possible that massage therapists and chiropractors are opportunistically referring patients for screening if they identify a skin lesion of concern. This finding suggests that massage therapists and chiropractors may present a further clinical opportunity for the early diagnosis of skin lesions. Initiatives aimed at educating this workforce to identify suspicious lesions and appropriate referral pathways may help to improve early skin cancer diagnosis and clinical outcomes.
Overall, rates for both breast and cervical screening were relatively low (approximately 50%) among our respondents. Even though rates were higher among those who consult CM practitioners, there may be an opportunity for CM practitioners to inquire about cancer screening as part of their routine care of female patients. The practice of asking patients about their cancer screening behavior may be particularly appropriate considering CM practitioners are commonly consulted by patients with chronic health conditions or other cancer risk factors. [bib_ref] Complementary medicine use by the Australian population: a critical mixed studies systematic..., Reid [/bib_ref] The daily routine care approaches and behaviors of CM practitioners concerning cancer screening are currently undetermined. Research is warranted to identify education needs and address any challenges the CM workforce may have concerning cancer screening initiatives, to maximize the public health opportunities of this large practitioner group.
# Limitations
The ALSWH is a comprehensive and well-respected source of epidemiological data, and a large number of participants and inclusion of the most important confounders within the regression models provide strength to the analyses reported here. There are, however, some limitations. The data collected were based on self-report and women may not have recalled all information correctly. We did not include conventional medical provider access and further study to examine relationships between conventional and CM provider utilization in relation to cancer screening would be valuable. Additionally, social desirability bias (respondents answer questions in a way they believe will please the researcher) cannot be ruled out. However, the opportunity to analyze CM consultation and cancer screening data from over 9000 women from a nationally representative survey, goes part way to countering these limitations.
# Conclusion
There appear to be various associations between women's consultation with each of a range of CM practitioner types and cancer screening uptake, and in summary, women who visit CM practitioners are generally more likely to participate in cancer screening initiatives. Further research investigating the utilization and perception of cancer screening by women who consult CM practitioners is essential in understanding how patients navigate cancer screening, and why some choose certain screening tools while others do not engage in cancer screening at all. Research is also required to understand the current and potential role that CM practitioners (can) have as public health advocates, recommending preventative health measures such as cancer screening. Considering the substantial prevalence of CM use worldwide, rich analyses of the relationship between CM practitioner use and cancer screening choice will help ensure optimal uptake of cancer screening and effective, timely care for all cancer patients.
[table] Table 1: The Bivariate Association Between Consulting Complementary Medicine Practitioners and Cancer Screening Behaviors, in 9151 Australian Women Aged 62 to 67 Years. a [/table]
[table] Table 2: Output From the Logistic Regression Models Showing the Association Between Consulting Health Care Practitioners and Cancer Screening Behaviors, in 9151 Australian Women Aged 62 to 67 Years. [/table]
|
Cytochrome P450 aromatase expression in human seminoma
Background:The enzyme cytochrome P450 aromatase, catalysing the conversion of androgens into estrogens, has been detected in normal human testicular cells suggesting a physiological role of local estrogen biosynthesis on spermatogenesis control. Estrogens, regulating cell growth and apoptosis, can also be involved in tumorigenesis process, but the possible link between estrogens and testicular neoplastic process is, up to now, scarcely known. This study examined aromatase expression in human seminoma, which is the most common germ cell tumour of the testis.Methods:The tumour-bearing testes were obtained from 20 patients with classic seminoma undergoing to therapeutic orchidectomy. Paraffin embedded tissues were processed for immunohistochemistry using a mouse monoclonal antibody generated against human placental cytochrome P450 arom, as primary antibody, and a biotinylated goat-anti-mouse IgG, as secondary antibody. Furthermore, Western blot analysis of seminoma extracts was carried out.Results: Intense P450 arom immunoreactivity was observed in the seminoma cells and Western blot analysis confirmed the immunodetection. A strong immunostaining was also detected in cells of intratubular germ cell neoplasia (IGCN), adjacent to seminoma.Conclusion:The present study demonstrated, for the first time in human, aromatase expression in neoplastic cells of seminoma suggesting a relation between local estrogen biosynthesis and germ cell tumorigenesis. The P450 arom immunolocalization in the cells of IGCN, representing the common precursor of most germ cell tumors, seems to support these findings.
# Background
Somatic and germ cell tumors are typical testicular neoplasms which can affect young, adult and elderly men. In the last years, testis tumor management has improved by new diagnostic and surgical tools, as well as by innovative therapy options to preserve fertility against cytotoxic treatments [bib_ref] Testicular sperm extraction in azoospermic patients with gonadal germ cell tumors prior..., Schrader [/bib_ref]. However, pathogenesis of male gonad malignancies is often still unknown.
Somatic and germ cells of normal human testis have revealed the expression of estrogens receptor-β and /or aromatase, which is the microsomial enzyme catalysing the conversion of androgens into estrogens [bib_ref] Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis, Simpson [/bib_ref] [bib_ref] Human testicular aromatase: Immunocytochemical and biochemical studies, Inkster [/bib_ref] [bib_ref] Estrogen regulation of testicular function, Akingbemi [/bib_ref] [bib_ref] Localization of oestrogen receptors α and β in human testis, Makinen [/bib_ref] [bib_ref] ERβ1 and ERβ2 splice variant (ERβcx/β2) are expressed in distinct cell populations..., Saunders [/bib_ref] [bib_ref] Development and validation of a new monoclonal antibody to mammalian aromatase, Turner [/bib_ref] [bib_ref] Differential distribution of splice variants of estrogen receptor beta in human testicular..., Aschim [/bib_ref]. Therefore, testicular cells are considered targets and in situ biosynthesis sites of estrogens, suggesting a physiological role of these hormones in steroidogenesis control and sperm maturation.
Estrogens are also known to regulate cell growth and apoptosis, therefore they can be involved in tumorigenesis process [bib_ref] Estrogen induced apoptosis by inhibition of the erythroid transcription factor GATA-1, Blobel [/bib_ref] [bib_ref] Estrogen inhibits paclitaxel induced apoptosis via the phosphorylation of apoptosis signal regulating..., Mabuchi [/bib_ref] but, up to now, their possible link with testicular neoplastic process is scarcely known.
Germ cell tumors, comprising seminoma and non-seminoma, are the most common malignancies among human male aged 15-40 years. Aim of this work was to investigate aromatase expression in seminoma which represents approximatively the 40% of these testicular neoplasms.
# Methods
## Samples and histopathological studies
The archival cases (collected during the last 3 years) were provided by the Pathologic Anatomy Unit (Annunziata Hospital). Tumoral testicular tissues were obtained from 20 patients (ages from 27 to 38 years) with unilateral classic seminoma undergoing to therapeutic orchidectomy. Normal testicular tissues were obtained from 2 patients showing testes with a granulomatous lesion. The ethical committee members of the University of Calabria approved the investigation programme.
Tissues were fixed in formalin (4%), dehydrated with a series of ethanol solutions and paraffin-embedded. Paraffin sections, 5 µm thick, were mounted on slides precoated with polylysine, deparaffinized and rehydrated (7-8 serial sections for each sample). These sections were used for morphological and immunohistochemical analyses.
Histopathological studies were carried out by Haematoxylin-Eosin staining. Seminoma samples were also investigated by placental-like alkaline phosphatase (PLAP) immunohistochemistry (monoclonal mouse anti human PLAP, clone 8A9, DAKO-Cytomation, Milan, Italy) Furthermore, 50 µm thick [bib_ref] Extraction and analysis of diagnostically useful proteins from formalin-fixed, paraffin-embedded tissue sections, Ikeda [/bib_ref] paraffin serial sections of seminoma specimens were cut for protein extraction and mounted on glass slides.
## P450 arom immunohistochemistry
Immunohistochemistry was performed after heat-mediated antigen retrieval [bib_ref] Antigen retrieval in formalin fixed, paraffin-embedded tissues: an enhancement method for, Shi [/bib_ref]. Hydrogen peroxide (3% in distillate water) was used, for 30 minutes, to inhibit endogenous peroxidase activity while normal goat serum (10%) was utilised, for 30 minutes, to block the non-specific binding sites. P450 arom immunodetection was carried out using a monoclonal mouse anti-human cytochrome P450 aromatase (MCA 2077, Serotec, Oxford,UK) (1:50 in TBS) at 4°C overnight. Then, a biotinylated horse-anti-mouse IgG (Vector Laboratories, CA, USA) was applied (1:600 in TBS) for 1 hour at RT, followed by the avidin-biotin-horseradish peroxidase com-plex (ABC/HRP) (Vector, Laboratories, CA, USA). Immunoreactivity was visualized by using the diaminobenzidine chromogen (DAB) (Zymed Laboratories, CA, USA) and, finally, sections were counterstained with haematoxylin. The primary antibody was replaced by normal rabbit serum in negative control, while the absorption control has utilised a primary antibody preabsorbed with an excess (5 nmol/ml) of purified human placental aromatase protein (4°C for 48 hours). Rat ovary sections were used as positive control. 6-7 serial sections were processed for each sample.
## Protein extraction
Protein extraction from formalin-fixed paraffin-embedded sections has been carried out according to Ikeda [bib_ref] Extraction and analysis of diagnostically useful proteins from formalin-fixed, paraffin-embedded tissue sections, Ikeda [/bib_ref]. Briefly, 50 µm testis sections from seminoma regions were deparaffinized in xylene, reydrated in graded ethanol, immersed in distilled water, and air dried. Then, the neoplastic area was recovered from the glass slides, further it was cut into small pieces and placed in Eppendorf tubes. Two hundred µl of RIPA buffer, pH 7,6 (1 M NaH 2 PO 4 , 10 mM Na 2 HPO 4 ,154 mM NaCl, 1% Triton X-100, 12 mM C 24 H 39 O 4 Na, 0.2% NaN 3 , 0.95 mM NaF, 2 mM PMSF, 50 mg/ml aprotinin, 50 mM leupeptin) (Sigma Chemical, Saint Louis, MO,USA) containing 0.2% SDS, was added to each tube and the contents were incubated at 100°C for 20 minutes, followed by incubation at 60°C for 2 hours. After incubation, tissue lysates were centrifuged at 15.000 × g for 20 minutes at 4°C and the supernatants were stored at -80°C until biochemical analysis.
# Western blot analysis
Tissue lysates were quantified using Bradford protein assay reagent [bib_ref] A rapid and sensitive method for the quantization of microgram quantities of..., Bradford [/bib_ref]. Equal amounts of protein (50 µg) were boiled for 5 minutes, separated under denaturing conditions by SDS-PAGE on 10% polyacrylamide Tris-glycine gels and electroblotted to nitrocellulose membrane. Nonspecific sites were blocked with 5% non fat dry milk in 0.2% Tween-20 in Tris-buffered saline (TBS-T) for 1 hour at room temperature and then probed with a rabbit polyclonal antiserum generated against human P450 aromatase(Hauptman-Woodward Medical Research Institute, Buffalo, NY, USA) (1:750). After extensive washings (three times for 15 minutes each time in TBS-T), a goat-antirabbit (1:7000) horseradish peroxidase-conjugated antibody (Vector Laboratories, CA, USA) was added for 1 hour at 22°C. Blots were again washed three times for 15 minutes in TBS-T and the bound of secondary antibody was located with the ECL Plus Western blotting detection system according to the manufacturer's instructions. Each membrane was exposed to the film for 2 minutes. P450 arom protein, isolated from human placenta, was used as positive control. Negative control was prepared using tissue lysate, where P450 arom was previously removed by preincubation with P450 arom antibody (1
# Results
## Histopathological study
Tumoral regions of all specimens have revealed the presence of classic seminoma with a uniform population of large neoplastic cells and extensive leukocytic infiltration [fig_ref] Figure 1: Morphology and P450 arom immunoreactivity of tumoral region in human testis with... [/fig_ref]. All seminoma specimens showed tumor stage I.
The region adjacent to seminoma (60% of samples) has shown the presence of abnormal seminiferous tubules with decreased tubular diameters and lacking spermatogenesis, which were identified as intratubular germ cell neoplasia (IGCN). These intratubular lesions were characterised by the basal proliferation of undifferentiated atypical enlarged germ cells, with big nuclei, and by the presence of Sertoli cells in luminal displacement.
The IGCN diagnosis was confirmed by PLAP immunohistochemistry. In fact, cytoplasmic and membranous dark staining have revealed the PLAP immunopositivity in the malignant cells aligned along the basal portion of seminiferous tubules [fig_ref] Figure 2: P450 arom immunoreactivity in testicular region adjacent to seminoma and in controls... [/fig_ref] , indicating their germ cell origin.
Morphology and P450 arom immunoreactivity of tumoral region in human testis with seminoma Furthermore, normal testes have displayed typical normal seminiferous tubules, with active spermatogenesis.
## P450 arom immunolocalization
P450 arom immunoreactivity in testicular cells was detected as cytoplasm staining while the nuclei were immunonegative.
In tumoral regions, neoplastic cell cords have revealed a strong immunoreactivity, while the surrounding abundant lymphocytes and connective cells were all immunonegative. The immunostaining pattern was similar in all the 20 examined samples and pictures in the figures show representative specimens [fig_ref] Figure 1: Morphology and P450 arom immunoreactivity of tumoral region in human testis with... [/fig_ref]. In all the 12 testis samples with intratubular germ cell neoplasia, an intense immunostaining has been observed in the cells of IGCN, adjacent to seminoma, either in the abnormal germ cells or in Sertoli cells [fig_ref] Figure 2: P450 arom immunoreactivity in testicular region adjacent to seminoma and in controls... [/fig_ref]. Leydig cells, clustered in interstitial tissue among IGCN, were strongly immunoreactive, representing an internal positive control (data not shown). Furthermore, the 2 normal testes have revealed an intense immunoreactivity in Leydig cells, while tubular compartments were all completely immunonegative [fig_ref] Figure 2: P450 arom immunoreactivity in testicular region adjacent to seminoma and in controls... [/fig_ref].
Negative (data not shown) and absorption controls (insert of [fig_ref] Figure 1: Morphology and P450 arom immunoreactivity of tumoral region in human testis with... [/fig_ref] were all immunonegative. As expected, ovarian tissue has shown a strong immunoreactivity in luteal cells (positive control) [fig_ref] Figure 2: P450 arom immunoreactivity in testicular region adjacent to seminoma and in controls... [/fig_ref].
# Western blot analysis
Lysates of all the 20 formalin-fixed paraffin-embedded testis sections from seminoma tumoral regions have been submitted to Western blot analysis. All specimens have shown a single band corresponding to a molecular mass of 55 kDa, which is the correct size of human P450 arom protein (3 representative samples are shown in . The band migrated at the same mobility as purified human placental P450 arom protein which was used as positive control, while the immunoreactive P450 arom band was lacking in the negative control
# Discussion
The present study has demonstrated, for the first time, aromatase expression in neoplastic cells of human seminoma, which is the most common testicular germ cell tumor of men in reproductive age.
Aromatase detection in Leydig cells, Sertoli cells and elongated spermatids of normal human testis has suggested the involvement of local estrogen production in gonadal physiology [bib_ref] Human testicular aromatase: Immunocytochemical and biochemical studies, Inkster [/bib_ref] [bib_ref] Estrogen regulation of testicular function, Akingbemi [/bib_ref] [bib_ref] Development and validation of a new monoclonal antibody to mammalian aromatase, Turner [/bib_ref]. At the same time, the mitogenic estrogen action could be related to the testicular tumor etiology, as hypothesized for human breast, ovarian, endometrial and epathic malignancies [bib_ref] Intratumoral aromatase in human breast, endometrial and ovarian malignancies, Sasano [/bib_ref] [bib_ref] Localized aberrant expression of cytochrome P450 aromatase in primary and metastatic malignant..., Harada [/bib_ref].
Epidemiological studies have indicated a higher incidence of testicular cancers after prenatal estrogen exposure [bib_ref] Estrogen exposure during gestation and risk of testicular cancer, Depue [/bib_ref] [bib_ref] Sugar L: Pre-natal and peri-natal exposure and risk of testicular germ cell..., Weir [/bib_ref] [bib_ref] Cancer risk in men exposed in utero to diethylstilbestrol, Strohsnitter [/bib_ref] [bib_ref] How valid is the prenatal estrogen excess hypothesis of testicular germ cell..., Dieckmann [/bib_ref]. Furthermore, a link between exposure to estrogens/ estrogen mimics and testicular cancer risk has been hypothesized [bib_ref] Occupational exposure to polyvinyl chloride as a risk factor for testicular cancer..., Hardell [/bib_ref] [bib_ref] Estrogen receptor-β expression in human testicular germ cell tumor, Pais [/bib_ref].
Few studies have investigated aromatase in testicular tumors. Among somatic testis tumors, a large cell calcify-Immunoblot of aromatase from seminoma extracts Immunoblot of aromatase from seminoma extracts. Lane 1: purified P450 arom protein from human placenta (positive control). Lane 2: negative control prepared as described in Material and Methods. Lanes 3, 4, 5: 55 kDa P450 arom immunoreactive bands from lysates of three different seminoma extracts. Numbers on the left correspond to molecular weights of marker proteins.
P450 arom immunoreactivity in testicular region adjacent to seminoma and in controls A: Intense aromatase immunos-taining in IGCN cells ing Sertoli cell tumorand Sertoli cell tumors of Peutz-Jeghers syndrome have shown an enhanced expression of this enzyme [bib_ref] An aromatase-producing sex-cord tumor resulting in prepubertal gynecomastia, Coen [/bib_ref] [bib_ref] Endocrine disorders associated with inappropriately high aromatase expression, Bulun [/bib_ref]. Furthermore, the single Nakazumi's report on testicular germ cell tumors has not detected aromatase expression in neoplastic cells of seminomas and non-seminomas but it revealed the enzyme presence only in stromal cells of non seminomas [bib_ref] Estrogen metabolism and impaired spermatogenesis in germ cell tumors of the testis, Nakazumi [/bib_ref].
The present study has shown aromatase immunolocalization in seminoma cells and Western blot analysis has confirmed this result. Our data disagree with Nakazumi's reports, but probably, the recent antibodies, used in our investigation, have improved aromatase immunodetection. In addition, our findings agree to a very recent work demonstrating the aromatase expression in human testicular seminoma cell line, JKT-1 [bib_ref] Estrogen-induced growth inhibition of human seminoma cells expressing estrogen receptor β and..., Roger [/bib_ref].
Furthermore, a strong P450 arom immunoreactivity has been observed inside the abnormal tubules of IGCN, which represents the common precursor of most germ cell tumours. Invasion of the tubular wall by malignant germ cells or intratubular proliferation of tumour cells are the two theories proposed to understand the switch from IGCN to testicular neoplasia [bib_ref] From carcinoma in situ to testicular germ cell tumor, Donner [/bib_ref]. However, to date, the mechanisms of invasive tumour development from pre-invasive lesion are still to be elucidated. P450 arom presence in malignant germ cells and in Sertoli cells of IGCN is a new finding which supports aromatase expression during seminoma development and suggests a possible role of local estrogen biosynthesis in tumorigenesis process. The aromatase absence in germ and Sertoli cells inside seminiferous tubules of normal testes supports this hypothesis. Concerning normal testes, aromatase detection in Leydig cells agrees with previous reports [bib_ref] Human testicular aromatase: Immunocytochemical and biochemical studies, Inkster [/bib_ref] [bib_ref] Development and validation of a new monoclonal antibody to mammalian aromatase, Turner [/bib_ref] , but its absence in seminiferous tubule cells could disagree with Turner's paper [bib_ref] Development and validation of a new monoclonal antibody to mammalian aromatase, Turner [/bib_ref] indicating a "possible" P450 arom expression in elongated spermatids and a "faint immunopositive reaction in Sertoli cells". However, the same authors claimed the immunostaining "less convincingly specific in human" than in marmoset and rat.
# Conclusion
Finally, the present study is the first report of aromatase expression in human testicular germ cells after malignant transformation, indicating that estrogens, from in situ aromatization, could act as autocrine mitogenic factors promoting neoplastic growth. These findings could represent a cue for further studies on estrogen involvement in testicular germ cell tumorigenesis and new pharmacological treatment. This work was supported by "Ministero dell'Universitàe della Ricerca Scientifica e Tecnologica" (Murst 60%) and by PRIN 2004 prot.n 0 0067227.
# Authors' contributions
CA: the author responsible for conception, design, analysis and interpretation of data as well as of drafting manuscript.
RV: the author responsible for performing the immunohistochemical expriments and participating in the analysis and interpretation of data.
RF: the author responsible for histoplathological diagnosis and sample collection AS and MD: the authors responsible for performing Western blot analysis and protein extraction as well as for participating to the interpretation of data.
ANS: the author responsible for a critical revision of the manuscript.
[fig] Figure 1: Morphology and P450 arom immunoreactivity of tumoral region in human testis with seminoma. A-B: Haematoxylin-eosin staining. C-D: Strong P450 arom immunoreactivity in cytoplasm of neoplastic cells (Nc) and unstained lymphocytes (L). Insert: absorption control. Scale bars: A, 20 µm; B-C, 12.5 µm; D, 5 µm. [/fig]
[fig] Figure 2: P450 arom immunoreactivity in testicular region adjacent to seminoma and in controls A: Intense aromatase immunostaining in IGCN cells. B: Placental-like alkaline phosphatase staining of IGCN basal cells C: Strong aromatase immunoreactivity of interstitial Leydig cells in normal testis (Lc) D: Intense immunostaining of luteal cells (Luc) in ovarian tissue Scale bars: A-B, 8 µm; C,12.5 µm; D, 5 µm. [/fig]
|
The correlation between serum selenium, zinc, and COVID-19 severity: an observational study
Background: Without an adequate immune response, SARS-CoV2 virus can simply spread throughout the body of the host. Two of the well-known immunonutrients are selenium (Se) and zinc (Zn). Se and Zn deficiency might lead to inflammation, oxidative stress, and viral entry into the cells by decreasing ACE-2 expression; three factors that are proposed to be involved in COVID-19 pathogenesis. Thus, in the current study we aimed at evaluating the correlation between serum Se and Zn status and COVID-19 severity.Methods: Eighty-four COVID-19 patients were enrolled in this observational study. Patients were diagnosed based on an infectious disease specialist diagnosis, using WHO interim guidance and the recommendations of the Iranian National Committee of Covid-19. The patients with acute respiratory tract infection symptoms were checked for compatibility of chest computed tomography (CT) scan results with that of Covid-19 and Real-time polymerase chain reaction (RT-PCR) for corona virus infection. The severity of Covid-19 was categorized into three groups (mild, moderate, and severe) using CDC criteria. Serum Zn and Se level of all subjects was measured. The severity of the disease was determined only once at the onset of disease.Results: According to the results of linear regression test, there was a significant association between Zn and Se level and COVID-19 severity (β = − 0.28, P-value = 0.01 for Se; β = − 0.26, P-value = 0.02). However the significance disappeared after adjusting for confounding factors. Spearman correlation analysis showed a significant negative association between serum Zn, Se and CRP level (r = − 0.35, P-value = 0.001 for Se; r = − 0.41, P-value < 0.001 for Zn).Conclusion:Results suggest that increasing levels of Se and Zn were accompanied by a decrease in serum CRP level. However, the significant association between Se, Zn, and disease severity was lost after adjusting for confounding factors.
# Introduction
COVID-19 is a multi-organ disease that correlates with heightened intensive care support and a high morbidity rate. Decreased immunity is a significant risk factor for viral respiratory infections. A proper diet and good nutritional status are important elements for an optimal immune response to prevent infection. Thus, a poor diet and nutritional deficiencies will increase the disease burden. Two of the most important immunonutrients are selenium (Se) and zinc (Zn).
The biological effects of Se derive from 25 selenoproteins, the most well-known of which are selenoprotein P, iodothyronine deiodinase, thioredoxin reductase, and glutathione peroxidase (GpX). For example, a decrease in the GpX-1 level will lead to an increase in the production of reactive oxygen species (ROS), activation of NF-κB transcription, and an increase in oxidative stress and cell apoptosis. Se is a well-known inhibitor of NF-kB and appears to decrease NF-kB-induced apoptosis and induce cytokine storms related to severe COVID-19. Recent studies have shown that the levels of selenoprotein P and GpX worldwide are borderline or below-optimal, even in many European countries. A study in China found that the mean hair selenium level was high in a Chinese city with a high COVID-19 recovery rate and low in cities with high COVID-19 mortality rates. Another population-based study and one cross-sectional study showed that Se deficiency correlated with the risk of COVID-19 mortality.
Up to 30% of common colds are caused by coronavirus infections. Studies report that any decrease in the severity, duration and frequency of the common cold after Zn supplementation will depend on the Zn compound used, the dosage, and onset of use after symptom initiation. In critically ill patients, persistently low serum Zn levels have an inverse correlation with recurrent sepsis and mortality from Streptococcus pneumoniae.
Infections with coronaviruses are accompanied by ciliary dyskinesia that result in impairment of mucociliary clearance. Physiological concentrations of Zn positively affect the length and number of bronchial cilia, as well as the ciliary beat frequency, which will improve the mucociliary clearance of viruses. Ex-vivo studies showed that a low Zn level increased the permeability of the epithelium of the respiratory tractand Zn supplementation enhanced lung integrity by inducing the expression of tight junction proteins (i.e. ZO-1 and claudin-1). Moreover, high Zn levels protect the lung from damage caused by mechanical ventilation. Zn adequacy can decrease viral entry into the cells by decreasing ACE-2 expression, inhibiting fusion with the host membrane and suppressing the RNA-dependent RNA polymerase of the virus.
A low serum Zn level has been correlated with increased ROS and pro-inflammatory markers. COVID-19-induced coagulopathy caused by microangiopathic organ failure, venous thromboembolism, and atherosclerosis development, which are the principal causes of death in these patients. Zn deficiency could promote thrombocyte aggregation and coagulation by inducing ROS production in platelets. Leukocytosis and neutrophilia have been associated with a poor prognosis for COVID-19. Recovery from lymphopenia can improve clinical recovery. Zn supplementation could reverse lymphopenia in innate immune cells. Zn is also a necessary regulator of TLR-3 and TLR-4 induced signaling pathways.
The only previously published observational study on the relationship between Zn deficiency and COVID-19 output reported a near significant relationship between Zn deficiency and longer hospital stays, as well as a higher prevalence of respiratory distress syndrome []. Few studies have been published about the link between COVID-19 outcomes and serum levels of Se and Zn. To the best of our knowledge, no study has assessed the association between serum Zn and Se level and COVID-19 severity. The present study investigated the potential link between serum levels of Se and severity of COVID-19.
# Methods
This was an observational prospective study which evaluated 84 patients diagnosed with COVID-19 presenting to the emergency ward of Sina Hospital, which is affiliated with Tehran University of Medical Sciences (TUMS). The data were collected up to 1 September 2020. The study was approved by the ethics committee of the Tehran University of Medical Sciences (IR.TUMS.VCR. REC.1399.134).
Subjects were diagnosed by an infectious disease specialist using WHO interim guidance and the recommendations of the Iranian National Committee on COVID-19. Medical records were collected from the COVID-19 Registry (SHCo-19R) database of Sina Hospital. SHCo-19R is a prospective, ongoing, hospital-based registry of patients diagnosed with COVID-19 who presented at the emergency ward of Sina Hospital.
The patients were 18 years of age and older with acute respiratory tract infection symptoms (e.g. cough, fever, dyspnea) with no other etiology that fully explained the clinical presentation. The compatibility of computed tomography (CT) chest scan results with that of COVID-19 detection and accurate diagnosis of a coronavirus infection from real-time polymerase chain reaction (RT-PCR) was used to detect the accuracy of diagnosis among patients. Evaluation of the severity of the disease was performed using CDC criteria for mild (fever, cough, sore throat, malaise, headache, muscle pain, nausea, vomiting, diarrhea, loss of taste and smell without shortness of breath, dyspnea, abnormal chest imaging), moderate (evidence of lower respiratory disease during clinical assessment or imaging, saturation of oxygen SpO 2 ≥ 94% in room air at sea level at 5-6 days after infection), severe (SpO 2 < 94% in room air at sea level, ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO 2 /FiO 2 ) < 300 mm Hg, respiratory frequency > 30 breaths/min, or lung infiltrates > 50%. respiratory frequency within 24-48 h), and critical (septic shock, respiratory failure, and/or multiple organ dysfunction/ failure).
Based on the available clinical data and the CDC criteria, we applied a severity risk score model that included three groups. Patients manifesting mild and moderate symptoms were assigned to groups one and two, respectively. Patients with severe and critical symptoms were assigned to group three. Severity score at the time of admission was used for further evaluations. Abuse of alcohol or other substances, pregnancy or lactation, and renal and kidney failure were considered as exclusion criteria.
## Study measurements
The present study collected data on the demographic characteristics of age, gender, past medical history, baseline clinical characteristics of the disease, onset of symptoms, and comorbidities. The results of RT-PCR, radiology and laboratory were included.
## Serum se, zn and biochemical assessment
The serum level of Se and Zn was measured only at the onset of study. Fasting blood samples (5 ml) were collected from all subjects in the control and case groups at the time of admission and were centrifuged within 30 min of collection. The serum samples were kept at − 20 °C.
Serum Se and Zn were measured using flame atomic absorption spectrometry (FAAS) technique (Atomic absorption spectrophotometer AA-680; Shimadzu; Japan). Serum CRP levels were determined by the immunoturbidimetry method using a CRP reagent kit (Audit Diagnostics; Ireland). The ESR was detected using the Westergren method and D-dimer measurements were performed using a latex agglutination test (Biorex Fars; Iran). A human ferritin enzyme immunoassay test kit (Immunobiological Laboratories; Germany) was used to assess serum ferritin levels through enzyme-linked immunosorbent assay (ELISA). The troponin level was also determined using ELISA kits (Shanghai Crystal Day Biotech; China).
Complete blood counts and differential leukocyte counts were performed using a cell counter (Nihdon Kohden Celltac E; Japan). The prothrombin time (PT) and partial thromboplastin time (PTT) measurements were performed using the appropriate kits and the international normalized ratio (INR) was detected accordingly. The glucose level was done measured by the glucose oxidase method (intra 144 assay CV = 1.8%; Pars Azmoon; Iran). Serum creatinine levels were assessed using the photometric Jaffe method with commercially available kits (Pars Azmoon; Iran). The serum electrolyte levels for Ca, Na, and K were determined by the ion selective electrode principle. ALT and AST serum levels were measured using a laboratory test kit (Pars Lab; Iran). The level of LDH was also assessed using an LDH assay kit (Takara; Japan). The serum CPK activity was detected using an auto analyzer.
# Statistical analysis
Determination of sample size in this study was based on our recent experience with similar study design and no calculation for statistical power was conducted prior to the study onset. A Shapiro-wilk test was used to evaluate normality distribution of data. In normal distributed continuous data, one-way ANOVA was applied. In data which was not normally distributed, ANOVA test was performed on log transformed data. Games-Howell test was used for post-hoc analyses. The association between Se, Zn, and COVID-19 severity score was analyzed using linear regression for normally distributed data. Linear regression was performed on log transformed data for Skewed variables. The ANOVA and association results were also adjusted by confounding factors. The correlation between Se, Zn and CRP was assessed using Spearman correlation analysis. Values are stated as numbers, and percentages, means and standard deviations and median and interquartile range according to the type and the tests applied. P-value < 0.05 was considered as statistical significance level in all performed analysis. Analyses were carried out using SPSS 21. All tests were two-tailed.shows that patients with severe COVID-19 were significantly older (81 ± 7 year) compared to patients in the mild (51 ± 14 year) and moderate (59 ± 14 year) groups (p < 0.001). Of these, 29.4% of patients were in the mild group, 37% in the moderate group, and 63.2% in the severe group required mechanical ventilation (p = 0.051).
# Results
## Baseline characteristics
Additional file 1: the biochemical assessments of the COVID-19 patients according to severity. Serum level of CRP (P-value = 0.031), urea (P-value < 0.001), vitamin B12 (P-value = 0.001), Se (P-value = 0.01), INR (P-value < 0.001), and troponin (P-value = 0.007) were significantly different across severity group. Other biochemical parameters didn't differ significantly among studied groups.indicates serum Se levels between COVID-19 cases according to disease severity categories. Accordingly, the mean ± SD of serum Se were as follows: 47.07 ± 20.82 ng/ml, 47.36 ± 25.6 ng/ ml, 29.86 ± 11.48 ng/ml in the mild, moderate and severe disease group respectively. According to simple linear regression model, there was a significant negative association between serum Se level and COVID-19 severity (standardized coefficient = − 0.28, P-value = 0.01). The association did not remain significant after adjusting for potential confounding factors including age and log-transformed of urea, CRP, INR, vitamin B12, lymphocyte count, hemoglobin, creatinine, troponin, and d-dimer. Zn level was also significantly different across severity categories (standardized coefficient = − 0.26, P-value = 0.02). Similar to Se, after controlling the confounding effects (Inc. age and log-transformed of urea, CRP, INR, vitamin B12, lymphocyte count, hemoglobin, creatinine, troponin, and d-dimer) in multiple linear regression model, no significant association was observed between serum Zn level and COVID-19 severity. Comparison of logtransformed Se and Zn levels and COVID-19 severity was also presented in Figs. 1 and 2 using scatter plots. the studied COVID-19 cases. Both minerals presented a statistically significant negative correlation with the inflammatory marker (CRP) levels using Spearman correlation analysis (r = − 0.35, P-value = 0.001 for Se; r = − 0.41, P-value < 0.001 for Zn).
## Serum zn and se and covid 19 severity
## Serum concentrations of selenium, zinc and crp in covid-19 cases
## Figures 3 and 4 describe the correlation between serum concentrations of selenium, zinc and crp in
# Discussion
To the best of our knowledge, this is the first study to investigate the association between Se and Zn status and COVID-19 severity. In the current study, serum Se levels were significantly lower in patients with severe COVID-19 compared to patients with mild or moderate disease. A recent population-based study in 17 China cities have provided notable evidence about the relationship between cure rates from COVID-19 and Se status. They reported that, as the Se hair concentration in a population increased, the recovery rate from COVID-19 also increased.
The results of a clinical study in Germany showed a strong association between the serum Se level and COVID-19 outcomes in hospitalized patients, with 39% of survivors and 65% of the deceased having low serum Se levels. The results of another study were in agreement with our results by showing that, in a clinical trial, supplementation with selenium, magnesium and vitamin B12 reduced the need for oxygen and/or intensive care support . However, after adjusting for confounding factors including age, urea, CRP, INR, vitamin B12, lymphocyte count, hemoglobin, creatinine, troponin, and d-dimer, the significance disappeared. It can be concluded that the effect of confounding factors especially age are stronger than the effect of serum selenium level in predicting COVID-19 severity. More ever, in current study Se level was far below the normal range (70-150 ng/ml) across three severity group (47.07 ± 20.82 ng/ ml, 47.36 ± 25.6 ng/ml, 29.86 ± 11.48 ng/ml in the mild, moderate and severe disease group respectively). Se sufficiency is needed for observing its anti-inflammatory and immune-enhancing effect in COVID-19 patients.
We observed a significant relationship between Se and CRP level. In accordance with our results, preclinical and clinical studies of other inflammatory diseases have shown a relationship between serum CRP and Se level. In a study on 137 critically ill children, an inverse association was observed between CRP and serum Se level . A preclinical study on mice assessed the oxidative stress and expression of inflammatory markers of primary-cultured peritoneal macrophages in Se-deficient and control groups . Se deficiency was found to ameliorate the antioxidant capacity and triggered the accumulation of oxygen free radicals. Se deficiency also significantly promoted the expression of the inflammatory mediators of IL-12, iNOS, IL-1β, NF-κB, and PTGe. An increase in NF-κB expression was followed by the accumulation of oxygen free radicals, which hindered the phagocytic capacity of the macrophages.
GpX and thioredoxin reductase play a pivotal role in modulating the inflammatory response and mediating the regulation of T-cell activity. Se prevents oxidative damage to endothelial cells and preserves their function. One life-threatening feature of COVID-19 is its involvement in thrombotic events such as deep-vein and arterial thrombosis, large vessel clots, microvascular thrombosis, and pulmonary embolism, probably caused by endothelium dysfunction, platelet activation, and inflammation.
SARS-CoV-2 has been shown to enhance endotheliitis induced by the infection of endothelial cells and the inflammatory response of the host. COVID-19 also causes thrombocytopenia and can induce stroke, even in young patients. The formation of thromboxane A2 (TxA2) is a key element in platelet aggregation and activation that can lead to blood thrombosis/coagulation. Sodium selenite has been reported to have antiaggregating properties from the reduction of TxA2 formation.
COVID-19 is more prevalent and more severe in older patients. Low or marginal Se levels are more prevalent and have been reported to be associated with increased ICU admission rates. In Sweden, 71% of older adults admitted to the ICU were Se deficient. Supplementation of Se for the institutionalized elderly was found to reduce the infection rate significantly.
In current study, we observed a significant association between Zn level and COVID-19 severity. Similar to Se, the association did not remain significant following the adjustment of confounding factors. We observed a significant negative relationship between serum Zn and CRP level. A number of review articles have examined the relationship between Zn level and COVID-19and discussed the relationship between Zn deficiency and deep respiratory infections other than COVID-19. Their findings concluded that Zn deficiency could be linked to the risk of infection and severe complication of COVID-19.
The only study available about Zn levels in Covid-19 patients showed that it was significantly lower in COVID-19 patients than the healthy control subjects. Deficient patients had higher rates of hospital stay, acute respiratory distress syndrome, and mortality []. Finding no association between Zn level and disease severity, could be explained by the stronger effect of confounding factors. Also it could in part be explained by the very low Zn serum levels in studied subjects. Serum Zn level in the majority of studied population was below 70 mcg/ml that considered as the minimum of the normal serum range. Zn adequacy is required to provide its immune-enhancing, anti-inflammatory, and other protective effects in COVID-19 patients. Available evidence also suggests that Zn deficiency is highly prevalent in Iran secondary to high phytate intake.
Because of the low proportion of patients with mortality in the current study we failed to assess the correlation between the mortality from disease and serum levels of Zn and Se. Furthermore, due to the observational structure of the current study, we couldn't describe the cause and effect relationship between serum Zn and Se level and COVID-19. Hence, further prospective cohort studies are needed. Another limitation of the present study was that we failed to follow up and assess patients in different stages of disease as we only collected the serum Se and Zn levels once at the onset of admission to hospital. However, it is recommended to be considered for further research in this field of study. According to the existing methodological issue, the current study results should be considered cautiously. Moreover, further studies are need to explore whether Zn and Se supplementation can improve COVID-19 characteristics and if so, unfold the undiscovered cellular or molecular mechanisms regarding the alleviating impact of this minerals on COVID-19 symptoms. |
Dosing Recommendations for Concomitant Medications During 3D Anti-HCV Therapy
The development of direct-acting antiviral (DAA) agents has reinvigorated the treatment of hepatitis C virus infection. The availability of multiple DAA agents and drug combinations has enabled the transition to interferon-free therapy that is applicable to a broad range of patients. However, these DAA combinations are not without drug-drug interactions (DDIs). As every possible DDI permutation cannot be evaluated in a clinical study, guidance is needed for healthcare providers to avoid or minimize drug interaction risk. In this review, we evaluated the DDI potential of the novel three-DAA combination of ombitasvir, paritaprevir, ritonavir, and dasabuvir (the 3D regimen) with more than 200 drugs representing 19 therapeutic drug classes. Outcomes of these DDI studies were compared with the metabolism and elimination routes of prospective concomitant medications to develop mechanism-based and drug-specific guidance on interaction potential. This analysis revealed that the 3D regimen is compatible with many of the drugs that are commonly prescribed to patients with hepatitis C virus infection. Where interaction is possible, risk can be mitigated by paying careful attention to concomitant medications, adjusting drug dosage as needed, and monitoring patient response and/or clinical parameters.Key PointsThe three direct-acting antiviral combination of ombitasvir, paritaprevir, ritonavir, and dasabuvir (3D regimen) is a combination therapy that was recently approved for the treatment of genotype-1 chronic hepatitis C virus infection.Potential drug-drug interactions with the 3D regimen were identified by applying pharmacokinetic study data to known routes of metabolism and disposition of more than 200 prescription and over-the-counter drugs.The majority of concomitant medications assessed are compatible with 3D therapy. Where interaction is possible, guidance on dose adjustment and/or clinical monitoring are provided. 276 P. S. Badri et al. Paritaprevir P-gp, BCRP, OATP1B1/B3 P-gp, BCRP, OATP1B1/B3 Dasabuvir P-gp, BCRP P-gp, BCRP Ritonavir P-gp P-gp, BCRP BCRP breast cancer resistance protein, OATP1B1 organic anion transporting polypeptide 1B1, OATP1B3 organic anion transporting polypeptide 1B3, P-gp p-glycoprotein, 3D direct-acting antiviral agent combination of ombitasvir, paritaprevir, ritonavir, and dasabuvir a Findings are based on in vitro data [24] DDI Potential of 3D Regimen 277
# Introduction
The development of direct-acting antiviral agents (DAAs) has revolutionized the treatment of chronic hepatitis C virus (HCV) infection. In head-to-head comparisons, combination therapy with DAAs has proven to be more effective and better tolerated than interferon-based therapies in both treatment-naïve and treatment-experienced patients [bib_ref] Boceprevir for previously treated chronic HCV genotype 1 infection, Bacon [/bib_ref] [bib_ref] Telaprevir for retreatment of HCV infection, Zeuzem [/bib_ref] [bib_ref] Sofosbuvir for previously untreated chronic hepatitis C infection, Lawitz [/bib_ref] [bib_ref] Simeprevir increases rate of sustained virologic response among treatment-experienced patients with HCV..., Zeuzem [/bib_ref]. One such investigational combination includes ombitasvir, paritaprevir (identified as a lead compound by AbbVie, Inc., North Chicago, IL, USA, and Enanta Pharmaceuticals, Inc., Watertown, MA, USA), ritonavir, and dasabuvir, together known as the 3D regimen. Ombitasvir, paritaprevir, and dasabuvir combine unique antiviral mechanisms of action (nonstructural protein 5A inhibition, nonstructural protein 3/4A protease inhibition, and non-nucleoside nonstructural protein 5B polymerase inhibition, respectively). This potent threeclass combination approach has achieved high rates of sustained virologic response in a broad range of patients, including those with cirrhosis or those who have undergone liver transplant [bib_ref] ABT-450/r-ombitasvir and dasabuvir with ribavirin for hepatitis C with cirrhosis, Poordad [/bib_ref] [bib_ref] An interferon-free antiviral regimen for HCV after liver transplantation, Kwo [/bib_ref]. The antiviral activity of paritaprevir is boosted by its co-formulation with a low dose of ritonavir (i.e., 100 mg), facilitating the use of a lower dose of paritaprevir and once-daily dosing. Ritonavir is a strong inhibitor of cytochrome P450 (CYP) 3A4, a major enzyme involved in the metabolism of paritaprevir.
In pivotal clinical trials, the 3D regimen with ribavirin achieved sustained virologic response rates at 12 weeks (SVR12) of 94-100 % in treatment-naïve and treatmentexperienced non-cirrhotic patients with genotype-1 HCV and 93-100 % after 24 weeks of treatment in patients with genotype-1 HCV and cirrhosis, including prior null responders [bib_ref] ABT-450/r-ombitasvir and dasabuvir with ribavirin for hepatitis C with cirrhosis, Poordad [/bib_ref] [bib_ref] ABT-450, ritonavir, ombitasvir, and dasabuvir achieves 97 % and 100 % sustained..., Andreone [/bib_ref] [bib_ref] ABT-450/r-ombitasvir and dasabuvir with or without ribavirin for HCV, Ferenci [/bib_ref] [bib_ref] Treatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin, Feld [/bib_ref] [bib_ref] Retreatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin, Zeuzem [/bib_ref]. Additionally, in liver transplant recipients with recurrent HCV genotype-1 infection and no cirrhosis (Metavir BF2) at least 12 months after transplantation, 33 of 34 patients (97 %; 95 % confidence interval [CI] 85-100 %) who were treated with the 3D regimen plus ribavirin for 24 weeks achieved SVR12 [bib_ref] An interferon-free antiviral regimen for HCV after liver transplantation, Kwo [/bib_ref]. No graft rejection events occurred during the study. The 3D regimen was well tolerated when administered with or without ribavirin; treatment discontinuation rates were low and adverse events (AEs) were generally mild [bib_ref] ABT-450/r-ombitasvir and dasabuvir with ribavirin for hepatitis C with cirrhosis, Poordad [/bib_ref] [bib_ref] An interferon-free antiviral regimen for HCV after liver transplantation, Kwo [/bib_ref] [bib_ref] ABT-450, ritonavir, ombitasvir, and dasabuvir achieves 97 % and 100 % sustained..., Andreone [/bib_ref] [bib_ref] ABT-450/r-ombitasvir and dasabuvir with or without ribavirin for HCV, Ferenci [/bib_ref] [bib_ref] Treatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin, Feld [/bib_ref] [bib_ref] Retreatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin, Zeuzem [/bib_ref]. In subjects receiving 3D with ribavirin, the most commonly reported AEs (occurring in [10 % of subjects) were fatigue, nausea, pruritus, other skin reactions, insomnia, and asthenia. In subjects receiving 3D regimen without ribavirin, the most commonly reported AEs (occurring in C5 % of subjects) were nausea, pruritus, and insomnia. The safety profile of the 3D regimen was similar in patients with cirrhosis [bib_ref] ABT-450/r-ombitasvir and dasabuvir with ribavirin for hepatitis C with cirrhosis, Poordad [/bib_ref] or who were posttransplant [bib_ref] An interferon-free antiviral regimen for HCV after liver transplantation, Kwo [/bib_ref] to that of the overall population and no significant associations were found between ombitasvir, dasabuvir, and ritonavir exposures and AEs or laboratory abnormalities [bib_ref] Exposure-safety response relationship for paritaprevir/ ritonavir, ombitasvir, dasabuvir and ribavirin in hepatitis..., Lin [/bib_ref]. Exposure-safety analyses showed that increases in paritaprevir exposure of up to 2-fold are not predicted to meaningfully increase AEs or laboratory abnormalities of Grade 3 or greater [bib_ref] Exposure-safety response relationship for paritaprevir/ ritonavir, ombitasvir, dasabuvir and ribavirin in hepatitis..., Lin [/bib_ref].
Comparisons of 3D pharmacokinetics in subjects with hepatic impairment vs normal hepatic function demonstrated that DAA exposures were not significantly affected (\35 % change) in subjects with mild hepatic impairment (Child-Pugh A) and, hence, no dosage adjustment of 3D therapy is required for such patients. In patients with moderate hepatic impairment (Child-Pugh B), ombitasvir, ritonavir, and dasabuvir area under the plasma concentration-time curves (AUCs) decreased by 30 % or less, whereas paritaprevir AUC increased by 62 %. Because of these exposure changes, 3D therapy is not recommended in patients with moderate hepatic impairment. A more substantial effect on DAA exposures (54 % decrease in ombitasvir and 325 and 945 % increase in dasabuvir and paritaprevir, respectively) was observed in patients with severe hepatic impairment (Child-Pugh C). Therefore, 3D therapy is contraindicated in patients with severe hepatic impairment.
Pharmacokinetic study data indicate that the presence of mild, moderate, or severe renal impairment does not increase DAA exposures to a clinically significant degree (\50 % change); therefore, no dose adjustment is indicated for 3D therapy when given to patients with renal impairment [bib_ref] The pharmacokinetics and safety of the direct acting antiviral regimen of ABT-450/r...., Khatri [/bib_ref].
Food has been shown to significantly increase exposures for 3D components, by as much as [bib_ref] Human hepatic cytochrome P450 2C9 catalyzes the rate-limiting pathway of torsemide metabolism, Miners [/bib_ref] [bib_ref] CYP2D6 and CYP3A4 involvement in the primary oxidative metabolism of hydrocodone by..., Hutchinson [/bib_ref] , and 30 % for ombitasvir, paritaprevir, ritonavir, and dasabuvir, respectively. Fat and/or calorie content do not appear to influence the magnitude of interaction. In all phase I-III studies, 3D regimen components were administered with food. Given the magnitude of the effect of food, 3D therapy should be administered with food to maximize absorption.
As new therapeutic options for treating HCV become available, it will be important to assess their abilities to interact with established medications, particularly those drugs or drug classes that are commonplace among patients with HCV infection. Adverse drug reactions resulting from concomitantly administered medications are an ongoing concern for patients undergoing HCV treatment because of the heavy burden of polypharmacy that coincides with a high prevalence of comorbidities in this patient population [bib_ref] Clinical management of drug-drug interactions in HCV therapy: challenges and solutions, Burger [/bib_ref] [bib_ref] Pattern and associated factors of anti-hepatitis C virus treatment-induced adverse reactions, Ravi [/bib_ref] [bib_ref] The high comorbidity burden of the hepatitis C virus infected population in..., Louie [/bib_ref]. The interplay of therapies influences not only the safety profile (via the potential for increasing drug exposures) but also may reduce therapeutic efficacy in cases where metabolic enzyme induction decreases circulating drug concentrations. Mechanistic data have shown a propensity for all DAAs to act as substrates, inhibitors and/or inducers of metabolic enzymes, and transporters, which is the foundation for the observed drug-drug interactions (DDIs) [bib_ref] Drug-drug interactions during antiviral therapy for chronic hepatitis C, Kiser [/bib_ref] [bib_ref] Different interaction profiles of direct-acting anti-hepatitis C virus agents with human organic..., Furihata [/bib_ref].
In clinical practice, non-HCV medications that have the potential for interactions with HCV treatments are frequently prescribed to patients with chronic HCV infection [bib_ref] The clinical significance of drug-drug interactions in the era of direct-acting anti-viral..., Maasoumy [/bib_ref] [bib_ref] Medication use and medical comorbidity in patients with chronic hepatitis C from..., Lauffenburger [/bib_ref]. In a recent study, more than half of the medications that were commonly prescribed to a large representative sample of patients with chronic HCV infection (N = 53,461) were found to have interaction potential with telapravir and boceprevir. [bib_ref] Medication use and medical comorbidity in patients with chronic hepatitis C from..., Lauffenburger [/bib_ref] However, later-generation DAAs represent an improvement in DDI propensity compared with the first-generation protease inhibitors.
Pharmacokinetic studies are the ideal method by which to evaluate the DDI potential of the 3D regimen; however, such an undertaking is not feasible given the number of prescription and over-the-counter medications available and their various permutations. In recognition of these limitations, regulatory agencies have created recommendations regarding which drugs should be studied to interrogate potential enzymatic and transporter pathways that may lead to significant DDIs. With results from key DDI studies, data can then be extrapolated to other medications based on what is known about drug metabolism.
To provide clear guidance regarding the clinical use of the 3D regimen and facilitate prescription decisions, in this report, we have gathered information on the metabolism and disposition of more than 200 commonly used drugs. These data, together with DDI studies involving the 3D regimen, were used to develop recommendations for DDI evaluation and management. In addition, these recommendations were successfully implemented for co-medication management in the pivotal phase III clinical trials of the 3D regimen and are also being implemented in the ongoing phase IIIb studies.
## Metabolic characteristics of 3d regimen components
In vitro preclinical and human pharmacokinetic studies have demonstrated that several enzymes and transporters are involved in the disposition of 3D regimen components. All DAAs show minimal renal elimination. The presence of ombitasvir, paritaprevir, and dasabuvir in feces suggests possible involvement of biliary excretion for these drugs. Within the family of CYP isoenzymes, paritaprevir and ritonavir are chiefly metabolized by CYP3A, whereas dasabuvir metabolism occurs primarily through CYP2C8 with minor contribution by CYP3A [bib_ref] A mechanistic non-clinical assessment of drug-drug interactions (metabolism and transporters) with the..., Bow [/bib_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Ombitasvir is mainly metabolized by amide hydrolysis. The DAAs are not expected to inhibit/induce CYP enzymes, although ritonavir is a known strong inhibitor of CYP3A, and the 3D regimen has been shown to modestly induce CYP1A2 and CYP2C19 in vivo [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref] , which is attributed to the effects of ritonavir.
In general, marketed and in-pipeline DAAs attain sufficient hepatocellular concentrations to inhibit HCV replication because they are substrates of hepatic transporters. This property also explains why these HCV therapeutic agents may interact with multiple drugs.
Components of the 3D regimen act as both substrates and inhibitors of transport proteins. The results of in vitro studies indicate involvement of one or more 3D agents as substrates and/or inhibitors of p-glycoprotein (P-gp), breast cancer resistance protein (BCRP), and organic anion transporting polypeptides 1B1 and 1B3 (OATP1B1/B3). However, none of the 3D components are substrates or inhibitors of renal transport proteins, including organic anion transporters (OAT1, OAT3), organic cation transporters (OCT1, OCT2), or multi-drug and toxin extrusion proteins (MATE1 and MATE2K), at clinically relevant concentrations [bib_ref] A mechanistic non-clinical assessment of drug-drug interactions (metabolism and transporters) with the..., Bow [/bib_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref].
Ombitasvir, paritaprevir, and dasabuvir inhibit uridine diphosphate glucuronosyltransferase (UGT) 1A1 [bib_ref] A mechanistic non-clinical assessment of drug-drug interactions (metabolism and transporters) with the..., Bow [/bib_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] ; however, the 3D regimen is not expected to inhibit UGT1A4, UGT1A6, UGT1A9, and UGT2B7 at clinically relevant concentrations. Historically, in vivo data have shown that ritonavir causes UGT1A1 induction [bib_ref] The effect of atazanavir and atazanavir/ ritonavir on UDP-glucuronosyltransferase using lamotrigine as..., Burger [/bib_ref] ; however, based on drug interaction data with raltegravir (a UGT1A1 substrate), the net effect of the 3D regimen is UGT1A1 inhibition [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interactions of the direct acting antiviral regimen of ABT-450/r, ombitasvir and..., Khatri [/bib_ref].
## Ddi studies with the 3d regimen
Eighteen formal interaction studies in healthy volunteers have been performed to evaluate the DDI potential of the 3D regimen with selected drugs, also known as probes, based on their disposition via a metabolic enzyme and/or transporter pathway [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. In these scenarios, the 3D regimen functioned either as (i) the 'victim', wherein the pharmacokinetics of one or more 3D agents were influenced by the presence of another drug, or (ii) the 'perpetrator', in which the 3D regimen affected the pharmacokinetics of the probe. Selection of probes was derived from general industry guidance issued by the US Food and Drug Administration and European Medicines Agency. Results from these mechanistic studies form the foundation for recommendations in this paper on dose adjustments for a wide range of drugs in various classes based on similar pharmacokinetic pathways. A summary of the results of these evaluations is presented in [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref].
## Ddi potential of concomitantly administered medications by therapeutic class
Using the results of the mechanistic studies described above, the potential for interactions of the 3D regimen with other commonly administered medications were determined based on the agents' pharmacokinetic profiles. Route of elimination information and other relevant influences on metabolism were identified from the published literature and/or prescribing information for more than 200 drugs. The following section describes the results of this analysis parsed by drug class. The ensuing dosing recommendations are summarized in [fig_ref] Table 2: Drug-drug interaction [/fig_ref].
With the vast size of the pharmaceutical armamentarium, a review of all of the possible drug interactions within the confines of this report is not possible. Therefore, in an effort to simplify the identification and management of drugs that may influence or be influenced by the 3D regimen, we have created the algorithm for screening and dose adjustment shown in [fig_ref] Figure 2: Proposed algorithm for screening, dose adjustment, and monitoring of potential drug interactions... [/fig_ref].
## Analgesics/anti-inflammatory agents
Within the analgesic/anti-inflammatory class, distinctions can generally be made between the interactions of nonsteroidal anti-inflammatory drugs (NSAIDs)/COX-2selective inhibitors and opioids. Metabolism of many of the NSAIDs/COX-2 inhibitors predominantly involves CYP2C9 and/or members of the UGT family [bib_ref] Characterization of rat and human UDP-glucuronosyltransferases responsible for the in vitro glucuronidation..., King [/bib_ref] [bib_ref] Identification of human UDP-glucuronosyltransferase isoform(s) responsible for the glucuronidation of 2-(4-chlorophenyl)-5-(2-furyl)-4-oxazoleacetic acid..., Kaji [/bib_ref] [bib_ref] Determination of CYP2C9-catalyzed diclofenac 4 0 -hydroxylation by high-performance liquid chromatography, Crespi [/bib_ref] [bib_ref] Stereoselective glucuronidation and hydroxylation of etodolac by UGT1A9 and CYP2C9 in man, Tougou [/bib_ref] [bib_ref] S)oxazepam glucuronidation is inhibited by ketoprofen and other substrates of UGT2B7, Patel [/bib_ref] [bib_ref] Glucuronidation of nonsteroidal anti-inflammatory drugs: identifying the enzymes responsible in human liver..., Kuehl [/bib_ref] [bib_ref] Effects of the antifungals voriconazole and fluconazole on the pharmacokinetics of s-(?)-and..., Hynninen [/bib_ref] [bib_ref] Cytochrome P450 2C9 catalyzes indomethacin Odemethylation in human liver microsomes, Nakajima [/bib_ref] [bib_ref] Nonsteroidal antiinflammatory drugs and phenols glucuronidation in Caco-2 cells: identification of the..., Sabolovic [/bib_ref] [bib_ref] Involvement of multiple cytochrome P450 isoforms in naproxen O-demethylation, Tracy [/bib_ref]. The lack of interaction between the model CYP2C9 substrate, warfarin, and the 3D regimen indicates that no dose adjustment would be required for drugs predominantly metabolized via CYP2C9 [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Moreover, the only interaction potential for a UGT species was that of UGT1A1, as demonstrated by the DDI study with raltegravir as the model substrate [bib_ref] A mechanistic non-clinical assessment of drug-drug interactions (metabolism and transporters) with the..., Bow [/bib_ref]. The involvement of multiple UGTs in NSAID glucuronidation likely minimizes of B20 % is considered to be not clinically significant. a No significant change was observed in exposures for norcyclobenzaprine, the major metabolite of cyclobenzaprine. BCRP breast cancer resistance protein, CYP cytochrome P450, OAT organic anion transporter, OATP organic anion transporting polypeptide, P-gp pglycoprotein, UGT uridine 5 0 -diphospho-glucuronosyltransferase. Data sources: [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref] [bib_ref] Drug-drug interactions of the direct acting antiviral regimen of ABT-450/r, ombitasvir and..., Khatri [/bib_ref] [bib_ref] Drug-drug interactions of omeprazole with the HCV direct acting antiviral (DAA) combination..., Polepally [/bib_ref] Carbama-zepine Phenytoin Phenobarbital any effect that the 3D regimen may have on UGT1A1based catalysis by providing alternate routes of metabolism [bib_ref] Glucuronidation of nonsteroidal anti-inflammatory drugs: identifying the enzymes responsible in human liver..., Kuehl [/bib_ref]. Notable exceptions to the general metabolism/elimination pattern for NSAIDs include acetaminophen, for which a 3D DDI study demonstrated minimal pharmacokinetic interaction (\20 % change in exposure) during concomitant therapy [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref] ; aspirin, which undergoes deacetylation and subsequent glucuronidation [bib_ref] Glucuronidation of the aspirin metabolite salicylic acid by expressed UDP-glucuronosyltransferases and human..., Kuehl [/bib_ref] ; ketorolac, which has a large renal excretion component [bib_ref] Paracetamol and ketorolac pharmacokinetics and metabolism at delivery and during postpartum, Allegaert [/bib_ref] ; and meloxicam, for which CYP3A4 plays a role [bib_ref] Meloxicam: a reappraisal of pharmacokinetics, efficacy and safety, Gates [/bib_ref]. The 3D regimen-mediated inhibition of CYP3A4 may increase meloxicam exposure; therefore, a dose reduction for meloxicam should be considered when it is co-administered with the 3D regimen. [fig_ref] Table 2: Drug-drug interaction [/fig_ref] provides a list of analgesics and anti-inflammatory agents that can be used with the 3D regimen without any dose adjustment. Contribution of CYP3A4 to the metabolism of most opioids (e.g., fentanyl, hydrocodone, hydromorphone, meperidine/pethidine, oxycodone, and tramadol) [bib_ref] Fentanyl metabolism by human hepatic and intestinal cytochrome P450 3A4: implications for..., Labroo [/bib_ref] [bib_ref] Pharmacogenomics as molecular autopsy for forensic toxicology: genotyping cytochrome P450 3A4*1B and..., Jin [/bib_ref] [bib_ref] CYP2D6 and CYP3A4 involvement in the primary oxidative metabolism of hydrocodone by..., Hutchinson [/bib_ref] [bib_ref] CYP2B6, CYP3A4, and CYP2C19 are responsible for the in vitro N-demethylation of..., Ramirez [/bib_ref] [bib_ref] Exposure to oral oxycodone is increased by concomitant inhibition of CYP2D6 and..., Gronlund [/bib_ref] [bib_ref] Effect of inhibition of cytochrome P450 enzymes 2D6 and 3A4 on the..., Gronlund [/bib_ref] [bib_ref] Effect of the inhibition of CYP3A4 or CYP2D6 on the pharmacokinetics and..., Kummer [/bib_ref] [bib_ref] Identification of cytochrome P-450 isoforms responsible for cis-tramadol metabolism in human liver..., Subrahmanyam [/bib_ref] indicates the potential for an interaction with the 3D regimen that would lead to increased drug exposure of the opioids. Hydrocodone AUC increased by 90 % when given with the 3D regimen in a DDI study [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref]. A 50 % reduction in hydrocodone dose is, therefore, recommended when administered with the 3D regimen. The other listed opioids should be used with caution and dose reduction is recommended when given concomitantly with the 3D regimen. Buprenorphine, another opioid, is a CYP3A substrate but no dose adjustment is recommended based on a modest interaction with the 3D regimen and the lack of effect on pharmacodynamic parameters [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. In a DDI study with the 3D regimen, no interaction was observed with naloxone [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] ; consequently, no dose adjustment is needed during co-administration. Codeine, morphine, and methadone are unlikely to require dose adjustment when given with 3D therapy because CYP3A4 is not the only or the primary
## Antidepressants
Drug-drug interactions with the 3D regimen are not expected for most antidepressants. The 3D regimen showed no significant interaction with the antidepressants duloxetine (a CYP2D6/CYP1A2 substrate) or escitalopram (a CYP2C19/CYP3A4 substrate) in DDI studies [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref].
Although the 3D regimen would be anticipated to decrease escitalopram exposure through CYP2C19 induction, the effects are likely offset by CYP3A4 inhibition. These results can be extrapolated to citalopram, which is a racemic mixture of escitalopram and (R)-citalopram. In addition, the 100-mg dose of ritonavir that is part of the 3D regimen is not expected to interact with substrates of CYP2D6 based on data with ritonavir 200 mg and the HIV protease inhibitor lopinavir. Thus, fluoxetine and paroxetine are unlikely to exhibit pharmacokinetic interactions with the 3D regimen, as they are also substrates of CYP2D6. No DDIs are expected for milnacipran because it is primarily excreted renally and shows minimal CYP450-related metabolism. Standard dosing for these antidepressants are warranted when given with the 3D regimen.
Historically, in vivo studies suggest an interaction between bupropion (CYP2B6 substrate) and ritonavir, with long-term administration of ritonavir 200 mg daily reducing bupropion exposure by as much as 22 % [bib_ref] Dose-related reduction in bupropion plasma concentrations by ritonavir, Park [/bib_ref]. However, no clinically significant interaction between the 3D regimen and bupropion, a CYP2B6 substrate, is expected based on studies of methadone, another CYP2B6 substrate, and the low dose (100 mg) of ritonavir that is co-formulated with the 3D regimen. Hence, initial dose modification is not needed for bupropion when administered with the 3D regimen. Clinical monitoring is recommended, with dose adjustment based on clinical symptoms.
The metabolism of fluvoxamine has not been fully elucidated, but is believed to be predominantly mediated by CYP1A2, with minor involvement of CYP3A4 and CYP2C19. Fluvoxamine also functions as a moderate inhibitor of CYP3A4 and CYP2C19. As a potential substrate or inhibitor of CYPs, alterations in pharmacokinetics are possible with co-administration of fluvoxamine and the 3D regimen. Caution should be exercised in prescribing fluvoxamine for patients receiving 3D therapy, and dose adjustment of fluvoxamine may be considered, depending on the clinical response.
Dose reduction is recommended for mirtazapine, reboxetine [bib_ref] Cytochrome P-450-mediated metabolism of the individual enantiomers of the antidepressant agent reboxetine..., Wienkers [/bib_ref] , sertraline [bib_ref] Grapefruit juice alters plasma sertraline levels after single ingestion of sertraline in..., Ueda [/bib_ref] , trazodone, and venlafaxinewhen administered with the 3D regimen because of the potential for an increase in drug exposure resulting from the inhibitory effects of the 3D regimen on CYP3A4.
Because of the potential for decreased ombitasvir, paritaprevir, ritonavir, and dasabuvir exposures and commensurate loss of therapeutic efficacy, the herbal antidepressant St. John's Wort is contraindicated for use with the 3D regimen.
## Antacids/proton pump inhibitors
The influence of acid reduction on 3D regimen pharmacokinetics was evaluated in a DDI study with omeprazole. The maximum concentration (C max ) and AUC values for ombitasvir, paritaprevir, ritonavir, and dasabuvir were only minimally effected (\20 % change in exposure); however, a 38 % reduction in omeprazole exposure was observed [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] [bib_ref] Drug-drug interactions of omeprazole with the HCV direct acting antiviral (DAA) combination..., Polepally [/bib_ref]. The results of this study indicate that (i) acid reduction, per se, does not significantly alter the pharmacokinetics of the 3D regimen, and (ii) a dose increase should be considered for substrates of CYP2C19, if a decrease in clinical response is noted. Other proton pump inhibitors (listed in [fig_ref] Table 2: Drug-drug interaction [/fig_ref] , follow a similar metabolic pathwayand dose increases are recommended if clinically indicated. However, dose Proposed algorithm for screening, dose adjustment, and monitoring of potential drug interactions with the direct-acting antiviral agent combination of ombitasvir, paritaprevir, ritonavir, and dasabuvir (3D regimen). CYP cytochrome P450, OAT organic anion transporter, OATP organic anion transporting polypeptide, P-gp p-glycoprotein, UGT uridine 5 0 -diphospho-glucuronosyltransferase adjustment is not expected for histamine H 2 -receptor antagonists (cimetidine, famotidine, ranitidine), which primarily undergo renal elimination. In addition, in vitro chelation experiments with the 3D regimen indicate that calcium, magnesium, and aluminum-based antacids will not interact with the 3D regimen.
## Diuretics
Dose adjustment is not warranted for the majority of diuretics when given with the 3D regimen. Amiloride, chlorthalidone, and hydrochlorothiazide undergo minimal hepatic metabolism and, hence, are not expected to demonstrate a metabolic interaction with the 3D regimen. Triamterene primarily undergoes sulfonation and there is no evidence of overlapping metabolic or transporter pathways with that of the 3D regimen. The metabolism of torsemide predominantly involves CYP2C9, with a minor contribution from CYP2C8 [bib_ref] Human hepatic cytochrome P450 2C9 catalyzes the rate-limiting pathway of torsemide metabolism, Miners [/bib_ref] [bib_ref] The xenobiotic inhibitor profile of cytochrome P4502C8, Ong [/bib_ref]. As demonstrated by the warfarin/3D DDI study, the CYP2C9 substrate does not show significant interaction with the 3D regimen [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref].
Dose adjustment may be necessary for furosemide and xipamide. In a DDI study, concomitant administration of furosemide and the 3D regimen increased the C max of furosemide by 42 %, although minimal change (8 %) in AUC was observed [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. The exact mechanistic reason for the greater increase in furosemide C max vs AUC is unknown but UGTs, including UGT1A1, may be a contributing factor. Clinical monitoring of patients receiving furosemide is recommended, with a possible dose reduction by as much as 50 %. Because the involvement of UGT1A1 cannot be ruled out, clinical monitoring is also suggested for patients who receive concomitant xipamide, with dose adjustment if clinically indicated.
## Antiplatelet agents/anticoagulants
Antiplatelet agents and anticoagulants are highly diverse in disposition, although none of the pathways are anticipated to lead to clinically relevant DDIs with the 3D regimen. Mechanisms including deacetylation and glucuronidation by various UGT species (aspirin) [bib_ref] Glucuronidation of the aspirin metabolite salicylic acid by expressed UDP-glucuronosyltransferases and human..., Kuehl [/bib_ref] , desulfation and/or depolymerization (enoxaparin), renal excretion (fondaparinux), and CYP2C9/CYP1A2 metabolism (acenocoumerol, fluindione, warfarin) [bib_ref] A pharmacokinetic-pharmacodynamic model for predicting the impact of CYP2C9 and VKORC1 polymorphisms..., Verstuyft [/bib_ref] [bib_ref] Cytochrome P4502C9 is the principal catalyst of racemic acenocoumarol hydroxylation reactions in..., Thijssen [/bib_ref] are not anticipated to be influenced by or to affect the pharmacokinetics of the 3D regimen. Indeed, no significant pharmacokinetic interaction was observed in a DDI study wherein warfarin was given concomitantly with the 3D regimen [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Of note, although dose adjustment is not indicated for the vitamin K antagonists warfarin and acenocoumerol when given with the 3D regimen, appropriate international normalized ratio monitoring is advisable according to routine clinical practice.
No dose adjustment is necessary for phenprocoumon, a vitamin K antagonist, although careful international normalized ratio monitoring is recommended when given with the 3D regimen based on its pharmacokinetic profile and drug interaction data with other relevant medications [bib_ref] Drug interactions with phenprocoumon and the risk of serious haemorrhage: a nested..., Jobski [/bib_ref]. It is suspected that the impact of CYP3A4, which is involved in the metabolism of phenprocoumon, is mitigated by the role of CYP2C9 and the notable portion of the administered drug that is excreted unchanged [bib_ref] Metabolic fate of phenprocoumon in humans, Toon [/bib_ref].
CYP3A4 and/or P-gp inhibition with the 3D regimen has implications for the use of apixaban, rivaroxaban, and dabigatran. Apixaban labeling recommends that the dose be reduced to 2.5 mg twice daily when given in combination with strong inhibitors of CYP3A4 and P-gp, such as ritonavir. Rivaroxaban is contraindicated for use with strong dual inhibitors of CYP3A and P-gp such as ketoconazole and ritonavirand, hence, should not be used with the 3D regimen. As a substrate of P-gp, dabigatran exposure is influenced by P-gp inhibition. Although levels of the P-gp substrate digoxin were not significantly affected by the 3D regimen in a DDI study [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] , exposures for dabigatran may increase as a result of intestinal P-gp inhibition by paritaprevir and ritonavir. Dabigatran should be used with caution with 3D therapy and appropriate international normalized ratio monitoring is advisable according to routine clinical practice.
An interaction with the purinergic receptor P2Y, G protein-coupled, 12 (P2Y12) antagonist clopidogrel is possible based on the involvement of CYP2C19 and CYP3A4/5 in its metabolism and bioactivation [bib_ref] Identification of the significant involvement and mechanistic role of CYP3A4/5 in clopidogrel..., Zhu [/bib_ref]. While the requirement for dose adjustment of clopidogrel in patients who initiate 3D therapy can be based on clinical judgment, therapeutic drug level monitoring for clopidogrel does not appear to be routine clinical practice. Hence, if clopidogrel cannot be avoided or substituted, it should be used with caution, or consideration can be given to use of an alternative DAA regimen that does not have interaction potential with this medication.
Conversion of prasugrel, a newer generation P2Y12 receptor antagonist, to its active metabolite is primarily dependent on CYP3A4 and CYP2B6, with some contribution by CYP2C9 and CYP2C19. Treatment with 100 mg ritonavir has been shown to decrease the C max and AUC of the active metabolite of prasugrel by 45 and 38 %, respectively [bib_ref] Pharmacokinetic interaction between prasugrel and ritonavir in healthy volunteers, Ancrenaz [/bib_ref]. This effect was attributed to inhibition of CYP3A4 by ritonavir. Nevertheless, concomitant administration of ketoconazole, a potent inhibitor of CYP3A4 and CYP3A5, with prasugrel did not impact inhibition of platelet aggregation by prasugrel despite a 34-46 % decrease in the C max of the active metabolite of prasugrel. A similar magnitude of interaction to that observed with ritonavir or ketoconazole is expected with the 3D regimen. Dose alteration is, therefore, not needed for prasugrel when initially administered with the 3D regimen; however, monitoring is recommended and dose adjustment should be enacted, if clinically indicated.
## Antihypertensive agents
Beta-blockers can generally be divided into two categories: those that do not undergo extensive metabolism by multiple enzymes in the liver (e.g., atenolol, bisoprolol, nadolol, and sotalol)and those with a metabolism that depends primarily on CYP2D6 (betaxolol, carteolol, metoprolol, nebivolol, propranolol, and timolol) [bib_ref] Association of CYP2D6 and ADRB1 genes with hypotensive and antichronotropic action of..., Zateyshchikov [/bib_ref] [bib_ref] Metabolism of carteolol by cDNA-expressed human cytochrome P450, Kudo [/bib_ref] [bib_ref] Timolol metabolism in human liver microsomes is mediated principally by CYP2D6, Volotinen [/bib_ref]. An exception to this classification is acebutolol, which undergoes a high degree of first-pass metabolism in the liver, but which has not been found to be a substrate of CYP3A4, CYP2D6, CYP2C9, CYP1A2, or CYP2C19 [bib_ref] Evaluation of recombinant cytochrome P450 enzymes as an in vitro system for..., Stringer [/bib_ref]. In all cases, no interaction between members of the beta-blocker class and the 3D regimen is expected.
The metabolic processes and pathways involved in the disposition of angiotensin-converting enzyme inhibitors (e.g., hydrolysis, glucuronidation, dimerization, cyclization, and renal elimination) [bib_ref] Disposition of fosinopril sodium in healthy subjects, Singhvi [/bib_ref] [bib_ref] Disposition of zofenopril calcium in healthy subjects, Singhvi [/bib_ref] are generally not anticipated to be affected by concomitant administration of the 3D regimen. Thus, no dose adjustment is indicated for captopril, fosinopril, lisinopril, perindopril, quinopril, ramipril, or zofenopril. One exception to this overarching guidance is enalapril, which is an OATP1B1 substrate [bib_ref] Effect of organic anion-transporting polypeptide 1B1 (OATP1B1) polymorphism on the single-and multiple-dose..., Tian [/bib_ref]. The 3D regimen has been shown to increase exposures of OATP1B1/B3 substrates [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] and, hence, a reduction in enalapril dose is advised for patients who are being treated with the 3D regimen.
The importance of CYP3A to the metabolism of calcium channel blockers and the inhibitory effects of certain calcium channel blockers on CYP3A are potential sources of interaction when combined with the 3D regimen [bib_ref] Grapefruit juice and cimetidine inhibit stereoselective metabolism of nitrendipine in humans, Soons [/bib_ref] [bib_ref] Drug interactions with calcium channel blockers: possible involvement of metabolite-intermediate complexation with..., Ma [/bib_ref]. With this in mind, a DDI study was conducted between the 3D regimen and the calcium channel blocker, amlodipine. Concomitant administration of the 3D regimen increased amlodipine exposure by up to 157 % [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Consequently, a 50 % dose reduction is recommended for amlodipine when given with the 3D regimen. Based on the prescribing information regarding use with CYP3A4 inhibitors and the potential for interaction [bib_ref] Grapefruit juice and cimetidine inhibit stereoselective metabolism of nitrendipine in humans, Soons [/bib_ref] [bib_ref] Drug interactions with calcium channel blockers: possible involvement of metabolite-intermediate complexation with..., Ma [/bib_ref] , the lowest possible starting dose should be used for diltiazem, nicardipine, nifedipine, nitrendipine, and verapamil when prescribed to patients receiving the 3D regimen. Both felodipine and nisoldipine have demonstrated a substantial interaction with CYP3A inhibitors (e.g., grapefruit juice, ketoconazole), which increases drug exposure by several fold. Therefore, neither felodipine nor nisoldipine should be given in conjunction with the 3D regimen.
Involvement of OATP transporters has been implicated in the disposition of many angiotensin II receptor blocker family members [bib_ref] Impact of OATP transporters on pharmacokinetics, Kalliokoski [/bib_ref] [bib_ref] Losartan is a substrate of organic anion transporting polypeptide 2B1, Flynn [/bib_ref] [bib_ref] In vitro and in silico strategies to identify OATP1B1 inhibitors and predict..., Karlgren [/bib_ref] [bib_ref] OATP and MRP2-mediated hepatic uptake and biliary excretion of eprosartan in rat..., Sun [/bib_ref]. Because of the influence of the 3D regimen on OATP1B1/B3 substrates [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] , dose reduction is suggested for OATP-associated angiotensin II receptor blockers, including candesartan, eprosartan, irbesartan, losartan, olmesartan, telmisartan, and valsartan. Azilsartan disposition, in contrast, does not appear to involve OATP transporters and no other overlapping metabolic pathways with the 3D regimen have been identified. Dose adjustment is, therefore, not necessary when azilsartan is administered with the 3D regimen.
No overlapping metabolic or transporter pathways have been identified between the aldosterone antagonist spironolactone and the 3D regimen [bib_ref] Spironolactone: disposition, metabolism, pharmacodynamics, and bioavailability, Karim [/bib_ref]. Hence, no dose adjustment would be required.
## Antiarrhythmic agents
The pharmacokinetics of digoxin were only minimally affected when given as a single dose (0.5 mg) to patients who had achieved steady state on the 3D regimen, suggesting that no dose adjustment is necessary for this combination therapy [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Based on the known metabolic pathways, drug-drug interaction is also not expected for other antiarrhythmic drugs whose metabolism involves CYP2D6 (flecainide)or CYP2D6/CYP1A2 (mexiletine) . However, these antiarrhythmic agents have a narrow therapeutic index; therefore, caution and therapeutic drug monitoring is suggested as part of routine clinical practice when these drugs are administered with the 3D regimen.
The involvement of CYP3A4 in the metabolism of amiodarone, disopyramide, dronedarone [bib_ref] Clinical pharmacology of dronedarone: implications for the therapy of atrial fibrillation, Dorian [/bib_ref] , propafenone, and quinidine [141] signals a possible increase in exposure when given with the 3D regimen owing to inhibitory effects on CYP3A4. If these agents cannot be avoided or substituted, they should be used cautiously, with careful monitoring and consideration for dose reduction, or use of an alternative DAA regimen that does not have interaction potential with these medications.
## Antidiabetic agents
Drug interactions are unlikely with the 3D regimen and most antidiabetic agents. Processes involved in the disposition of most approved antidiabetic agents, including urinary excretion (acarbose, metformin, sitagliptin), hydrolysis (vildagliptin), and metabolism via CYP1A1 (pioglitazone), CYP2C8 (pioglitazone, rosiglitazone, sitagliptin), and/or CYP2C9 (glimepiride, glipizide, rosiglitazone, tolbutamide) [bib_ref] Potential CYP2C9-mediated drug-drug interactions in hospitalized type 2 diabetes mellitus patients treated..., Tirkkonen [/bib_ref] [bib_ref] Impact of CYP2C9 amino acid polymorphisms on glyburide kinetics and on the..., Kirchheiner [/bib_ref] [bib_ref] Tolbutamide, flurbiprofen, and losartan as probes of CYP2C9 activity in humans, Lee [/bib_ref] , are unlikely to affect the pharmacokinetics of the 3D regimen, nor is the 3D regimen likely to influence exposures to these drugs. In a DDI study of the 3D regimen with metformin, changes in metformin drug exposure during concomitant administration were not clinically significant (B23 % decrease) [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref].
The magnitude of CYP3A4 involvement in the metabolism of other antidiabetic agents influences the degree of interaction potential with the 3D regimen. Labeled guidance suggests a dose limitation to 2.5 mg once daily for saxagliptin, a substrate of CYP3A4, when given in conjunction with a strong CYP3A4 inhibitor, such as the 3D regimen. Dose reduction is recommended for repaglinide, a substrate of CYP2C8, CYP3A4, and OATP1B1, when used with the 3D regimen. Mechanistically, interaction is possible through both inhibition of CYP3A4 and OATP1B1 by the 3D regimen [bib_ref] A mechanistic non-clinical assessment of drug-drug interactions (metabolism and transporters) with the..., Bow [/bib_ref] ; DDI studies of the 3D regimen with OATP1B1/B3 substrates [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] and repaglinide with OATP1B1 inhibitors support a potential interaction [bib_ref] Effects of gemfibrozil and atorvastatin on the pharmacokinetics of repaglinide in relation..., Kalliokoski [/bib_ref] [bib_ref] Cyclosporine markedly raises the plasma concentrations of repaglinide, Kajosaari [/bib_ref]. Inhibition of OATP by the 3D regimen may also increase exposure to glibenclamide (glyburide); hence, dose reduction is recommended for glibenclamide when administered with the 3D regimen [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] [bib_ref] Elucidating rifampin's inducing and inhibiting effects on glyburide pharmacokinetics and blood glucose..., Zheng [/bib_ref]. In contrast, because CYP3A4 only plays a minor role in the metabolism of sitagliptin and the majority of the drug (79 %) is excreted unchanged in urine, the effect of 3D-mediated CYP3A4 inhibition on sitagliptin pharmacokinetics is expected to be minimal in patients with normal renal function. However, in patients with severe renal impairment or end-stage renal disease, sitagliptin exposure may increase when co-administered with the 3D regimen.
## Lipid-modifying therapies
As described above, studies of the 3D regimen with pravastatin or rosuvastatin revealed increases in exposure for both of these drugs [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. To accommodate these changes, it is recommended that pravastatin dose be reduced by half and that the rosuvastatin dose is limited to 5 or 10 mg when prescribed to patients receiving the 3D regimen. As substrates of OATP1B1 and/or OATP1B3 [bib_ref] Impact of OATP transporters on pharmacokinetics, Kalliokoski [/bib_ref] , the pharmacokinetics of atorvastatin, cerivastatin, ezetimibe, fluvastatin, and pitavastatin may also be influenced by the 3D regimen. Although CYP3A4 does play a role in the metabolism of the aforementioned statins [bib_ref] Impact of OATP transporters on pharmacokinetics, Kalliokoski [/bib_ref] , atorvastatin is the only one for which it is the dominant CYP species. Atorvastatin is, hence, not recommended with the 3D regimen. If a statin is required, clinicians should use the lowest available dose of cerivastatin, ezetimibe, fluvastatin, and pitavastatin, or switch to a reduced recommended doses of pravastatin/rosuvastatin when given concomitantly with the 3D regimen.
No overlap in metabolic or transporter pathways with the 3D regimen was identified for niacin, omega-3 fatty acids, and mipomersen; these agents can be administered with the 3D regimen without dose adjustment. Fenofibric acid, the pharmacologically active moiety of fenofibrate and choline fenofibrate, undergoes glucuronidation by several UGT species, including UGT1A1, UGT1A3, UGT1A9, and UGT2B7 [bib_ref] Effects of fibrates on metabolism of statins in human hepatocytes, Prueksaritanont [/bib_ref]. The contribution of UGT1A1 to this process is comparatively modest. Therefore, dose adjustment is not indicated, although monitoring is suggested for fenofibric acid.
As anion exchange resins, the bile acid sequestrants colestipol and colesevelam may bind to molecules other than their intended targets. This may lead to delayed or incomplete absorption of concomitantly administered oral medications. Consistent with prescribing guidance for bile acid sequestrants and the absorption window of the 3D regimen, it is recommended that the administration of bile acid sequestrants should be separated from the 3D regimen administration by at least 4 h.
Several lipid-modifying agents should not be given in conjunction with the 3D regimen, including gemfibrozil, lovastatin, and simvastatin. Gemfibrozil was shown to mediate a substantial increase (approximately ten-fold) in dasabuvir exposure when administered with paritaprevir/r and dasabuvir [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Recommendations for using lovastatin and simvastatin are based on the labeled contraindication for combination with CYP3A inhibitors because of an increased risk for myopathy, including rhabdomyolysis.
## A 1 -adrenergic blockers
The use of a 1 -adrenergic blockers have several restrictions based on their CYP3A4-dependent metabolism. Neither tamsulosin or alfuzosin should be given with the 3D regimen because of restrictions on their use with CYP3A inhibitors in their respective labels. However, doxazosin can be used with caution and dose reduction.
## B-adrenergic agonists
There are no overlapping metabolic/transporter pathways between albuterol (salbutamol) and 3D regimen components. Hence, a drug interaction is not expected between albuterol and the 3D regimen and no dose adjustment is necessary when administered together.
The metabolism of formoterol involves an array of enzymes, some of which (UGT1A1, CYP2C19) are known to be affected by the 3D regimen. Given the potential for altering drug pharmacokinetics, formoterol should be used with caution and dose reduction should be implemented if clinically indicated. However, concomitant use of salmeterol, a substrate of CYP3A4, is not recommended with the 3D regimen because of the potential for QTc prolongation, palpitation, and sinus tachycardia.
## Hypnotic/sedative agents
Recommendations for co-administration of hypnotic/ sedative agents differ by individual drugs. CYP3A4 plays a dominant role in the metabolism of many hypnotic/ sedative agents, including that of alprazolam, clonazepam, eszopiclone, estazolam, flunitrazepam, prazepam, quazepam, and zaleplon and zopiclone [bib_ref] Identification of human cytochrome P450 enzymes involved in the formation of 4-hydroxyestazolam..., Miura [/bib_ref] [bib_ref] Metabolic activation of benzodiazepines by CYP3A4, Mizuno [/bib_ref]. Because of the inhibitory effect of the 3D regimen on this enzyme, dose reduction is generally recommended when these agents are administered to patients receiving 3D treatment. The recommended dose adjustment is supported by a DDI study with alprazolam in which the drug's AUC increased by 34 % when administered with the 3D regimen [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref].
Notable exceptions to the dose reduction guidance include diazepam, zolpidem, triazolam, and midazolam. Changes in exposures for diazepam, a substrate of both CYP3A4 and CYP2C19, were modest (22 % decrease in AUC and 18 % increase in C max ) when coadministered with the 3D regimen in a DDI study [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref]. However, the C max of nordiazepam, the major metabolite of diazepam, increased by 10 % and the AUC for a dosing interval and AUC from 0 to infinity (AUC inf ) decreased by 3 and 44 %, respectively. Because of its long half-life (137 h), AUC inf was not reliably estimated when diazepam was dosed alone. Given these results, diazepam dose can be increased, if clinically indicated, although pre-emptive dose adjustments are not required. When given concomitantly with the 3D regimen in a DDI study, zolpidem pharmacokinetics were not appreciably affected; therefore, dose reduction is not indicated [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Triazolam and oral midazolam, in contrast, are contraindicated for use with medications that impair CYP3A function, and should, therefore, not be given with the 3D regimen. Although the effect of the 3D regimen on midazolam exposure is expected to be significantly less with parenteral administration, close monitoring for respiratory depression and/or prolonged sedation are recommended and dose adjustment may be considered.
Another potential influence on DDI risk with hypnotics/ sedatives is glucuronidation involving different members of the UGT family. Delayed plasma clearance of lorazepam resulting from inhibition of glucuronidation has been observed in clinical studies [bib_ref] The effect of valproic acid on drug and steroid glucuronidation by expressed..., Ethell [/bib_ref]. Given the influence of the 3D regimen on UGT1A1 substrates [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref] , lorazepam dose reduction is suggested. Dose reduction, however, is not required for oxazepam and temazepam owing to the minimal involvement of UGT1A1 in their glucuronidation [bib_ref] UDP-glucuronosyltransferases and clinical drug-drug interactions, Kiang [/bib_ref]. In general, clinical monitoring should be considered for all hypnotic/sedative agents, irrespective of the dose adjustment recommendation.
## Antiepileptic agents
Methods of disposition for many antiepileptic drugs, including renal elimination (felbamate, gabapentin, topiramate, vigabatrin)and hydrolysis (levetiracetam), are generally not anticipated to be influenced by the 3D regimen. Dose alteration is also not anticipated for sodium valproate, whose metabolism involves multiple UGT and CYP species [bib_ref] Effect of aging on glucuronidation of valproic acid in human liver microsomes..., Argikar [/bib_ref] ; however, therapeutic drug monitoring is recommended when initiating coadministration.
Exposure to drugs that are metabolized by CYP3A4, including tiagabineand zonisamide, may increase with concomitant 3D therapy. These drugs should be used with caution and dose reduction should be considered. Dose alteration and patient monitoring may also be required for lamotrigine owing to UGT1A1 induction by ritonavir [bib_ref] The effect of atazanavir and atazanavir/ ritonavir on UDP-glucuronosyltransferase using lamotrigine as..., Burger [/bib_ref].
Carbamazepine, phenobarbital, and phenytoin are inducers of CYP3A4 [bib_ref] Clinically relevant drug interactions with antiepileptic drugs, Perucca [/bib_ref]. In a DDI study with the 3D regimen, carbamazepine appreciably decreased DAA exposure [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Because of the potential for loss of therapeutic efficacy, carbamazepine, phenobarbital, and phenytoin are contraindicated for use with the 3D regimen.
## Steroids
Co-administration of ritonavir with corticosteroids that are metabolized by CYP3A4 (e.g., budesonide, dexamethasone, fluticasone, methylprednisolone, mometasone, prednisone, and triamcinolone) increases steroid concentrations and can lead to iatrogenic Cushing syndrome and adrenal suppression [bib_ref] Metabolic pathways of inhaled glucocorticoids by the CYP3A enzymes, Moore [/bib_ref] [bib_ref] Triamcinolone and ritonavir leading to drug-induced Cushing syndrome and adrenal suppression: description..., Schwarze-Zander [/bib_ref]. Caution should be exercised and lowest doses should be used when considering use of these agents with the 3D regimen. In particular, budesonide and fluticasone, especially if prescribed for long-term use, should only be initiated if the potential benefit of treatment outweighs the risk of systemic corticosteroid effects.
Beclomethasone dipropionate appears to be less prone to interactions with ritonavir. Direct combination of ritonavir with beclomethasone increased exposures of its metabolite, beclomethasone-17-monopropionate (17-BMP), but not to a level that was considered to be clinically significant [bib_ref] Influence of low-dose ritonavir with and without darunavir on the pharmacokinetics and..., Boyd [/bib_ref]. Interestingly, 17-BMP exposure was not appreciably increased when darunavir and ritonavir were both administered with beclomethasone. Beclomethasone is an acceptable alternative inhaled/intranasal corticosteroid option for patients receiving the 3D regimen. The pharmacokinetic characteristics of ciclesonide generate lower systemic exposure than other inhaled/intranasal corticosteroids [bib_ref] Relevance of pharmacokinetics and pharmacodynamics of inhaled corticosteroids to asthma, Derendorf [/bib_ref]. Ciclesonide may be given without dose adjustment to patients treated with the 3D regimen, with the caveat that monitoring for steroid-related side effects should be ongoing.
The use of topical corticosteroid creams or lotions with the 3D regimen is permissible with attention to factors that can increase systemic absorption such as use over large body surface areas or prolonged use, as described in prescribing information.
## Antibiotics
Beta-lactams and fluroquinolones can be given with the 3D regimen without dose adjustment because they are predominantly excreted unchanged in urine or bileand there are no apparent overlapping metabolic/transporter pathways between these antibiotics and the 3D regimen. The possible involvement of P-gp in azithromycin disposition [bib_ref] A study of the pharmacokinetics of azithromycin and nelfinavir when coadministered in..., Amsden [/bib_ref] is not believed to have an appreciable impact on drug pharmacokinetics, given the lack of significant interaction detected with the 3D regimen and digoxin, a model P-gp substrate [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Clarithromycin is a strong inhibitor of CYP3A4, through which an interaction with the 3D regimen is possible. Nonetheless, a pharmacokinetic evaluation of concomitant ritonavir and clarithromycin determined that no dose adjustment was required for either agent when co-administered to patients with normal renal function [bib_ref] Pharmacokinetic interaction between ritonavir and clarithromycin, Ouellet [/bib_ref]. Clarithromycin dose reduction is, however, recommended for patients with moderate or severe renal impairment.
Clinically significant interactions were not observed between sulfamethoxazole/trimethoprim and the 3D regimen in a DDI study. The moderate increase (33 %) in dasabuvir exposure is attributable to the weak CYP2C8 inhibitory effects of trimethoprim [bib_ref] Drug-drug interactions of commonly used medications with direct acting antiviral HCV combination..., Polepally [/bib_ref]. Exposures to sulfamethoxazole and trimethoprim modestly increased (\25 %). The magnitude of these changes does not indicate that dose adjustment is needed for either the 3D regimen or sulfamethoxazole/trimethoprim during coadministration.
Metabolic pathways for fusidic acid are not currently known, but interaction with drugs metabolized by CYP3A4 is suspected. It is recommended that fusidic acid not be given to patients who are receiving a drug or drugs that are metabolized by CYP3A4. As such, fusidic acid is contraindicated for use with the 3D regimen.
## Antifungals
A potential for interaction with the 3D regimen exists for several antifungal agents, including ketoconazole, itraconazole, posaconazole, and voriconazole, by virtue of their metabolism by or effects on CYP3A4. In a DDI study, the AUC of ketoconazole (a CYP3A4/P-gp inhibitor and CYP3A4 substrate) increased by 117 % when a single dose of the 3D regimen was added to steady-state ketoconazole 400 mg once daily [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Although ombitasvir, paritaprevir, ritonavir, and dasabuvir exposures also increased, none of the changes were considered clinically significant. Similar increases are possible with itraconazole, another CYP3A4/ P-gp inhibitor and CYP3A4 substrate. As a strong inhibitor of CYP3A4, caution is advised when posaconazole is administered with drugs metabolized by CYP3A4, including ritonavir. Alterations in voriconazole exposure are also possible owing to its metabolism by CYP2C19 and CYP3A4. Unlike the azoles, echinocandins such as capsofungin, micafungin, and anidulafungin do not have primary interactions with CYP enzymes or P-gp; therefore, a drug interaction is not expected between these drugs and the 3D regimen.
In summary, in patients being treated with the 3D regimen, daily doses of greater than 200 mg are not recommended for either ketoconazole or itraconazole, and use of a lower posaconazole dose is advised. Co-administration of voriconazole with the 3D regimen should be avoided unless an assessment of the benefit-to-risk balance justifies its use. Echinocandins can be administered with the 3D regimen without any dose modification.
## Oral contraceptives
Interactions with the 3D regimen have been evaluated for progestin-only contraceptives and combined oral contraceptives. Minimal effects on drug exposures were observed when norethindrone 0.35 mg (a progestin-only contraceptive) was administered with the 3D regimen. However, use of 3D therapy with norgestimate (250 lg) plus ethinyl estradiol (35 lg) significantly increased the levels of norgestimate metabolites and appreciably decreased paritaprevir, ritonavir, and dasabuvir exposures. Moreover, asymptomatic alanine transaminase elevations were observed when the 3D regimen was combined with norgestimate (250 lg) plus ethinyl estradiol (35 lg) or norethindrone (0.40 mg) plus ethinyl estradiol (35 lg) [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. Given the potential for alanine transaminase elevations in women who use ethinyl estradiol-containing medications such as combined oral contraceptives, contraceptive patches, or contraceptive vaginal rings, these medications must be discontinued at least 2 weeks prior to starting therapy with the 3D regimen. Alternative methods of contraception (e.g., progestin-only contraception or nonhormonal methods) are recommended during 3D therapy. Ethinyl estradiol-containing medications can be restarted approximately 2 weeks after completion of 3D treatment [bib_ref] Drug-drug interaction profile of the all-oral anti-hepatitis C virus regimen of paritaprevir/ritonavir,..., Menon [/bib_ref]. If ethinyl estradiol-containing oral contraceptives cannot be avoided or substituted, consideration can be given to use of an alternative DAA regimen that does not have interaction potential with these medications.
## Phosphodiesterase type 5 inhibitors
The predominant route of metabolism for phosphodiesterase type 5 (PDE5) inhibitors is through CYP3A family members. As such, PDE5 inhibitor clearance is expected to be reduced in the presence of CYP3A inhibitors, including the 3D regimen. Guidance for use of PDE5 inhibitors with CYP3A inhibitors is dependent on the therapeutic indication. In the absence of data for the 3D regimen, limitations and contraindications suggested for low-dose ritonavir can be applied.
PDE5 inhibitors are used both in the treatment of erectile dysfunction and pulmonary arterial hypertension, at different doses. For the treatment of erectile dysfunction in patients receiving low-dose ritonavir, doses of sildenafil, tadalafil, and vardenafil should be limited to 25 mg every 48 h, 10 mg every 72 h, or 2.5 mg every 72 h, respectively. Avanafil should not be prescribed for patients receiving 3D, as no safe and effective dose has been determined when used with low-dose ritonavir therapy. In patients with pulmonary arterial hypertension, combining sildenafil with low-dose ritonavir is contraindicated because of a lack of an established safe and effective dose. For patients with pulmonary arterial hypertension, consideration can be given to use of an alternative DAA regimen that does not have interaction potential with sildenafil. Tadalafil and vardenafil 20 mg once daily may be initiated after at least 1 week of low-dose ritonavir treatment. The dose may be increased to 40 mg once daily if tolerated and clinically indicated. Increased monitoring for AEs is recommended during concomitant PDE5 inhibitor/3D therapy.
## Thyroid replacement therapy
Many of the drugs that are commonly used by patients with HCV are covered within the classes discussed above. Based on a recent evaluation of prescription patterns in patients with chronic HCV [bib_ref] Medication use and medical comorbidity in patients with chronic hepatitis C from..., Lauffenburger [/bib_ref] , one additional medication that is often prescribed to patients with HCV is worthy of note. Levothyroxine sodium is a frequent concomitant medication among patients with chronic HCV owing to the elevated prevalence of hypothyroidism in this population [bib_ref] Thyroid disorders in chronic hepatitis C, Antonelli [/bib_ref]. Levothyroxine is a sensitive substrate that is thought to be metabolized by UGT1A1 [bib_ref] Drug-drug interactions for UDP-glucuronosyltransferase substrates: a pharmacokinetic explanation for typically observed low..., Williams [/bib_ref] without the confounding influence of other UGTs or CYPs. Increases in thyroxine concentration have been observed with UGT1A1 inhibitors such as indinavir [bib_ref] Interaction between levothyroxine and indinavir in a patient with HIV infection, Lanzafame [/bib_ref]. Based on results of the 3D/UGT1A1 substrate DDI study [fig_ref] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir,... [/fig_ref] , concomitant use of levothyroxine with the 3D regimen may increase concentrations of levothyroxine. Hence, thyroid-stimulating hormone levels should be carefully monitored in patients receiving levothyroxine and 3D and dose adjustment implemented based on the clinical response.
Two classes of agents that are absent from this discussion are medications used in the treatment of HIV and immunosuppressants used to prevent transplant rejection (e.g., cyclosporine and tacrolimus). The results of ongoing and completed studies [bib_ref] Pharmacokinetics and dose recommendations for cyclosporine and tacrolimus when coadministered with ABT-450,..., Badri [/bib_ref] with these agents will be presented in a separate review.
## Clinical experience with the 3d regimen
The 3D regimen has been evaluated with or without ribavirin in six phase III clinical trials that enrolled more than 2300 patients infected with genotype-1 HCV [bib_ref] ABT-450/r-ombitasvir and dasabuvir with ribavirin for hepatitis C with cirrhosis, Poordad [/bib_ref] [bib_ref] ABT-450, ritonavir, ombitasvir, and dasabuvir achieves 97 % and 100 % sustained..., Andreone [/bib_ref] [bib_ref] ABT-450/r-ombitasvir and dasabuvir with or without ribavirin for HCV, Ferenci [/bib_ref] [bib_ref] Treatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin, Feld [/bib_ref] [bib_ref] Retreatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin, Zeuzem [/bib_ref]. Patients were allowed to continue most of their co-medications while receiving study drug treatment, although dose adjustment was permissible and recommended. More than 1200 co-medications belonging to 15 drug classes and 19 enzyme/transporter inhibitor/inducer categories were given concomitantly with the 3D regimen in these studies [bib_ref] Drug-drug interactions with direct acting antiviral combination therapy of ABT-450/r, ombitasvir and..., Menon [/bib_ref] [bib_ref] Pharmacokinetics of paritaprevir, ombitasvir, dasabuvir, ritonavir and ribavirin in subjects with HCV..., Mensing [/bib_ref] [bib_ref] Effect of comedications on paritaprevir, ritonavir, ombitasvir, dasabuvir and ribavirin pharmacokinetics, Polepally [/bib_ref]. Percentage-of-use metrics based on drug class are listed in [fig_ref] Table 2: Drug-drug interaction [/fig_ref]. Polypharmacy was commonplace, with approximately 1500 patients (65 %) receiving two or more concomitant medications from multiple drug classes/categories [bib_ref] Effect of comedications on paritaprevir, ritonavir, ombitasvir, dasabuvir and ribavirin pharmacokinetics, Polepally [/bib_ref]. Patient monitoring and dose adjustment, as necessary, were sufficient to manage any potential DDIs. The 3D regimen with or without ribavirin was well tolerated in treatment-naïve and treatment-experienced HCVinfected patients with and without cirrhosis; the rate of study drug discontinuation due to AEs was low (\1 %) and few serious AEs were reported.
# Conclusions
Our analysis suggests that the 3D regimen is compatible with many of the drugs that are commonly used by patients with HCV infection. Where an interaction is possible, risk can be mitigated by careful attention to concomitant medications, adjusting drug dosage as needed, and monitoring patient response and/or clinical parameters. As HCV 3D therapy is indicated for only 12-24 weeks, clinicians may choose to either suspend existing or new interacting medications or clinically manage the potential DDIs over the fixed, short-term, 3D regimen treatment duration. By using these strategies with the drug-specific guidance provided herein, the likelihood for adverse drug reactions can be lessened, thereby maximizing the opportunity for successful treatment of HCV infection. While these recommendations provide general guidelines based on known mechanism of disposition for the 3D regimen and various interaction medications, clinical judgment based on patient response should prevail during co-administration.
[fig] Figure 1: Formal interaction studies performed with the direct-acting antiviral agent combination of ombitasvir, paritaprevir, ritonavir, and dasabuvir (3D regimen) and their implications on dose recommendations for other drugs. A change in exposure (maximum concentration [C max ] or area under the plasma concentration-time curve [AUC]) [/fig]
[fig] Figure 2: Proposed algorithm for screening, dose adjustment, and monitoring of potential drug interactions with the direct-acting antiviral agent combination of ombitasvir, paritaprevir, ritonavir, and dasabuvir (3D regimen). CYP cytochrome P450, OAT organic anion transporter, OATP organic anion transporting polypeptide, P-gp p-glycoprotein, UGT uridine 5 0 -diphospho-glucuronosyltransferase [/fig]
[table] Table 2: Drug-drug interaction (DDI) potential of commonly administered medications [/table]
|
ERK inhibitor ASN007 effectively overcomes acquired resistance to EGFR inhibitor in non‐small cell lung cancer
The emergence of acquired resistance limits the long-term efficacy of EGFR tyrosine kinase inhibitors (EGFR TKIs). Thus, development of effective strategies to overcome resistance to EGFR TKI is urgently needed. Multiple mechanisms to reactivate ERK signaling have been successfully demonstrated in acquired resistance models. We found that in EGFR mutant non-small cell lung cancer (NSCLC) patients, acquired resistance to EGFR TKIs was accompanied by increased activation of ERK. Increased ERK activation was also found in in vitro models of acquired EGFR TKI resistance. ASN007 is a potent selective ERK1/2 inhibitor with promising antitumor activity in cancers with BRAF and RAS mutations. ASN007 treatment impeded tumor cell growth and the cell cycle in EGFR TKI-resistant cells. In addition, combination treatment with ASN007 and EGFR TKIs significantly decreased the survival of resistant cells, enhanced induction of apoptosis, and effectively inhibited the growth of erlotinib-resistant xenografts, providing the preclinical rationale for testing combinations of ASN007 and EGFR TKIs in EGFR-mutated NSCLC patients. This study emphasizes the importance of targeting ERK signaling in maintaining the long-term benefits of EGFR TKIs by overcoming acquired resistance.
# Introduction
Although EGFR tyrosine kinase inhibitors (EGFR TKIs) significantly improve clinical outcomes in patients with nonsmall cell lung cancer (NSCLC) harboring EGFR mutations, such as exon 19 deletion and the L858R mutation, almost all mutant EGFR-positive NSCLC patients ultimately acquire resistance to EGFR TKIs after approximately 1-2 years [bib_ref] Overcoming drug resistance to receptor tyrosine kinase inhibitors: Learning from lung cancer, Kuwano [/bib_ref]. Therefore, overcoming acquired resistance is still crucial to improving therapeutic efficacy. The most common acquired resistance mechanism is acquisition of another resistant mutation in EGFR, such as T790M and C797S. In addition, resistance to EGFR TKIs can also be driven by constitutive activation of the MEK-ERK pathway, as demonstrated previously [bib_ref] Acquired resistance to AZD9291 as an upfront treatment is dependent on ERK..., Ku [/bib_ref] [bib_ref] CXCR7 reactivates ERK signaling to promote resistance to EGFR kinase inhibitors in..., Becker [/bib_ref] [bib_ref] Combined EGFR/MEK inhibition prevents the emergence of resistance in EGFR-mutant lung cancer, Tricker [/bib_ref] [bib_ref] Reactivation of ERK signaling causes resistance to EGFR kinase inhibitors, Ercan [/bib_ref] [bib_ref] Optimizing the sequence of anti-EGFR-targeted therapy in EGFRmutant lung cancer, Meador [/bib_ref] [bib_ref] Acquired resistance to the mutant-selective EGFR inhibitor AZD9291 is associated with increased..., Eberlein [/bib_ref].
MEK-ERK signaling is a key pathway downstream of EGFR and mediates EGFR-dependent regulation of cancer cell growth and survival. Previous studies have demonstrated that sustained ERK activation is involved in resistance to EGFR TKIs. Thus, targeting MEK-ERK signaling through either an MEK or ERK inhibitor can overcome acquired resistance to EGFR TKIs. Aberrant activation of ERK signaling was found in erlotinib-, gefitinib-, osimertinib-, and WZ4002-resistant NSCLC cells [bib_ref] Acquired resistance to AZD9291 as an upfront treatment is dependent on ERK..., Ku [/bib_ref] [bib_ref] Reactivation of ERK signaling causes resistance to EGFR kinase inhibitors, Ercan [/bib_ref] [bib_ref] Optimizing the sequence of anti-EGFR-targeted therapy in EGFRmutant lung cancer, Meador [/bib_ref] [bib_ref] Acquired resistance to the mutant-selective EGFR inhibitor AZD9291 is associated with increased..., Eberlein [/bib_ref] [bib_ref] Src mediates ERK reactivation in gefitinib resistance in non-small cell lung cancer, Ochi [/bib_ref] [bib_ref] Adaptive and acquired resistance to EGFR inhibitors converge on the MAPK pathway, Ma [/bib_ref] [bib_ref] ERK inhibition effectively overcomes acquired resistance of epidermal growth factor receptor-mutant non-small..., Li [/bib_ref] [bib_ref] MEK or ERK inhibition effectively abrogates emergence of acquired osimertinib resistance in..., Gu [/bib_ref] [bib_ref] Overcoming acquired resistance to AZD9291, a third-generation EGFR inhibitor, through modulation of..., Shi [/bib_ref]. Furthermore, in erlotinib-and gefitinib-resistant cells, the combination of EGFR TKI with an MEK inhibitor effectively inhibited tumor growth and impeded the development of resistance [bib_ref] Src mediates ERK reactivation in gefitinib resistance in non-small cell lung cancer, Ochi [/bib_ref] [bib_ref] Adaptive and acquired resistance to EGFR inhibitors converge on the MAPK pathway, Ma [/bib_ref]. Similarly, the combination of osimertinib with an MEK or ERK inhibitor synergistically induced cell death in osimertinib-resistant NSCLC cells [bib_ref] Acquired resistance to AZD9291 as an upfront treatment is dependent on ERK..., Ku [/bib_ref] [bib_ref] Acquired resistance to the mutant-selective EGFR inhibitor AZD9291 is associated with increased..., Eberlein [/bib_ref] [bib_ref] ERK inhibition effectively overcomes acquired resistance of epidermal growth factor receptor-mutant non-small..., Li [/bib_ref] [bib_ref] MEK or ERK inhibition effectively abrogates emergence of acquired osimertinib resistance in..., Gu [/bib_ref] [bib_ref] Overcoming acquired resistance to AZD9291, a third-generation EGFR inhibitor, through modulation of..., Shi [/bib_ref]. MEK inhibition using trametinib combined with WZ4002 was shown to delay the emergence of acquired resistance to WZ4002 in NSCLC [bib_ref] Combined EGFR/MEK inhibition prevents the emergence of resistance in EGFR-mutant lung cancer, Tricker [/bib_ref].
FOS-related antigen 1 (FRA1) is an ERK-dependent oncogenic transcription factor and a member of the AP-1 transcriptional factor superfamily. As ERK amplitude and duration both contribute to the induction and activation of FRA1, the expression and phosphorylation levels of FRA1 protein linearly reflect ERK activity [bib_ref] Linear integration of ERK activity predominates over persistence detection in Fra-1 regulation, Gillies [/bib_ref] [bib_ref] 2020) FRA1 contributes to MEK-ERK pathway-dependent PD-L1 upregulation by KRAS mutation in..., Lee [/bib_ref]. FRA1 is frequently upregulated in a wide variety of tumors and has important roles during consecutive stages of multistep tumor progression by promoting cell proliferation, inhibiting apoptosis and enhancing tumor angiogenesis. In addition, FRA1 may promote cancer progression by facilitating immune evasion through PD-L1 expression in high-risk, premalignant bronchial epithelial cells [bib_ref] 2020) FRA1 contributes to MEK-ERK pathway-dependent PD-L1 upregulation by KRAS mutation in..., Lee [/bib_ref].
ERK is immediately downstream of MEK and is important to many cellular processes. ERK is responsible for phosphorylating a broad range of substrates involved in cell proliferation, differentiation, and survival. Selective ERK inhibitors have been developed and used in clinical trials for the treatment of a variety of cancers such as melanoma, pancreatic cancer, and NSCLC [bib_ref] Targeting ERK1/2 protein-serine/threonine kinases in human cancers, Roskoski [/bib_ref]. However, given the negative feedback upregulation of MEK and limited activity of ERK inhibitors alone, current strategies include combination treatment with an MEK inhibitor to impede ERK activation. Furthermore, selective ERK inhibitors may be a promising strategy for minimizing toxicity and enhancing activity.
ASN007 is an oral ERK1/2 inhibitor; an open-label, doseescalation phase I study of ASN007 began in January 2018 and is still ongoing. However, to date, little is known about function of ASN007 in preclinical models. Here, we investigated whether ASN007 alone or combination can overcome acquired resistance to EGFR TKIs in NSCLC.
# Materials and methods
## Patient tissue samples and immunohistochemistry
Patients treated with EGFR TKIs (erlotinib, gefitinib, and afatinib) at Samsung Medical Center were retrospectively identified based on baseline and post-progression FFPE tissue availability. All procedures involving tumor specimens were reviewed and approved by the Institutional Review Board (IRB) of Samsung Medical Center (No. SMC 2010-04-039, 2011-10-054, 2013-08-113, and 2013-10-112), and written informed consent was provided by patients; in some cases, a waiver of consent was obtained. Paired tissue sections were obtained form 34 NSCLC patients and used for p-ERK1/2 staining. Immunohistochemistry was performed on 4-µm sections of formalin-fixed paraffin-embedded samples. Following deparaffinization and rehydration of the slides, antigen retrieval was performed using citrate buffer (pH 6.0). After endogenous peroxidase activity was blocked with 3 % hydrogen peroxide, sections were incubated with primary antibody for p-ERK1/2 (1:300, #4376; Cell Signaling Technology). Sections were further processed with horseradish peroxidase-conjugated secondary antibody and then developed with 3,3-diaminobenzidine. Finally, the slides were counterstained with hematoxylin. Slides were scanned with the Scanscope XT (Aperio Leica BioSystems) and analyzed with ImageJ software.
## Cell cultures and reagents
As previously described [bib_ref] Acquired resistance to AZD9291 as an upfront treatment is dependent on ERK..., Ku [/bib_ref] , an erlotinib-resistant cell line (PC9/ER) and osimertinib-resistant cell lines (PC9/OR and H1975/OR) have been established in our laboratory. Resistant cells contained original EGFR mutations, but had no additional EGFR mutations such as T790M or C797S. Cells were cultured in RPMI-1640 medium supplemented with 10 % FBS, penicillin (100 U/ml), and streptomycin (100 µg/ml) at 37°C in a humidified atmosphere containing 5 % CO 2 . ASN007 was provided by Asana BioSciences (Bridgewater, New Jersey). Erlotinib and osimertinib were obtained from Selleckchem. All drugs were dissolved in dimethyl sulfoxide (DMSO) at a 10 mM concentration and stored at -20°C until further use.
## Cell viability assay and colony formation assay
Cell viability was determined using a Cell Counting Kit (Dojindo Molecular Technologies) according to the manufacturer's instructions. Colony formation assay was used to measure long-term cell viability, Briefly, cells were seeded in 6well plates and allowed to attach overnight. Following the indicated treatment, drug containing medium was changed every 3 days. After 10-14 days, the colonies were fixed and stained with crystal violet.
## Western blotting and antibodies
Cells were lysed on ice in lysis buffer (50 mM Tris (pH 8.0), 150 mM NaCl, 1 % NP-40) supplemented with a protease and phosphatase inhibitor cocktail (Sigma). Equal amounts of protein were then subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes. After blocking in 5 % skim milk, membranes were incubated with primary antibodies to p-ERK1/2 (Thr202/ Tyr204). ERK1/2, p-FRA1(Ser265), FRA1, p-p90RSK (S380), RSK1, FoxM1, Aurora A, PLK1, Cyclin D1, CyclinB1, and PARP (Cell Signaling Technology) and β-actin (Santa Cruz Biotechnology). Membranes were then incubated with the appropriate second antibodies and developed using ECL (Pierce).
# Cell cycle analysis
Cell cycle analysis was performed after 24 h of treatment. Cells were fixed with ice-cold 70 % ethanol, stained with propidium iodide, and analyzed by flow cytometry (BD Biosciences).
## Xenograft studies
The protocol involving all procedures about animals was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) at Samsung Biomedical Research Institute (SBRI). They are in accordance with the relevant national and international guidelines. Six-week-old BALB/c female nude mice were injected subcutaneously with PC9/ER cells. When tumor size reached approximately 100 mm 3 , mice were randomly assigned to groups of 4-6 mice each. ASN007 was dissolved in 0.5 % methyl cellulose containing 0.1 % Tween-80 and given orally. To evaluated the synergistic effect of EGFR TKI and ASN007, each group of mice was dosed with vehicle, ASN007 (50 mg/kg/d), erlotinib (25 mg/kg/d), or a combination of both by oral gavage 5 days per week. Tumor volumes were determined using calipers and calculated using the following formula: V = (L x W 2 )/2 (L, length; W, width). Mice were monitored daily with humane endpoints including a tumor greater than 1,500 mm 3 , weight loss of over 15 % of body mass, vomiting or skin problems, or inability to ambulate or rise for food and water. These humane endpoints were not observed in any mouse. All efforts were made to alleviate suffering. Mice were euthanized by CO 2 inhalation at the end of the experiment. Tissues obtained after sacrifice were used for molecular analysis and Ki-67 (1:200, #9027; Cell Signaling Technology) staining.
# Statistical analysis
Data are presented as mean ± SEM. Statistical analyses were carried out using GraphPad Prism (GraphPad software). Correlation analysis was conducted using Pearson's correlation. Statistical evaluation was performed with a two-tailed Student's t-test, and P values < 0.05 were considered statistically significant.
# Results
## Acquired resistance to egfr tki is related to increased erk activation in nsclc tumor tissue
To determine whether increased ERK activation is related to EGFR-independent bypass resistance in NSCLC, we retrospectively analyzed 34 paired (pre-and post-treatment) biopsy samples from EGFR TKI treated patients. The treatment regimen in this cohort was erlotinib in 4 patients (11.8 %), gefitinib in 14 patients (41.2 %), and afatinib in 16 patients (47.1 %). Two patients (one treated with erlotinib and another with afatinib) received osimertinib as second-line treatment before re-biopsy. Consistent with a previous report [bib_ref] Adaptive and acquired resistance to EGFR inhibitors converge on the MAPK pathway, Ma [/bib_ref] , 26 (76.5 %) tumor biopsies collected at disease progression after EGFR TKIs treatment showed high level ERK phosphorylation on immunohistochemistry . Activated ERK level was significantly higher in post-treatment biopsy compared with baseline regardless of EGFR TKI regimen , suggesting that increased ERK activation may be associated with acquired resistance to EGFR TKI in EGFR-mutant NSCLC. However, the treatment period of EGFR TKI showed no correlation with ERK activation levels .
## Erk inhibitor asn007 inhibits cell proliferation in egfr tki-resistant nsclc cells
To evaluate the impact of ERK activation in acquired resistance to EGFR TKI, we examined the activation level of ERK and ERK substrates in both parental (PC9 and H1975) and EGFR TKI-resistant cells (erlotinib-resistance: PC9/ER, osimertinib-resistance: PC9/OR and H1975/OR) [fig_ref] Figure 2: ERK inhibitor ASN007 induces cell death in EGFR TKIresistant NSCLC cell lines [/fig_ref]. Each resistance cell line harbors genomic alterations involved in ERK activation (PC9/ER: NRAS Q61H, PC9/OR: HRAS G13R, H1975/OR: MAP2K1 K57E). The resistant cells showed increased ERK phosphorylation compared to parental cells [fig_ref] Figure 2: ERK inhibitor ASN007 induces cell death in EGFR TKIresistant NSCLC cell lines [/fig_ref]. Unlike evident phosphorylation of FRA1, p90RSK phosphorylation was rarely detected in PC9, PC9/ ER, and PC9/OR cells regardless of ERK phosphorylation status. To determine whether the survival of the resistant clones was the result of increased ERK activation, we treated ERK inhibitor, ASN007 to parent and resistant cells. ASN007 induced apoptotic cell death, as indicated by cleaved PARP and PAPR cleavage was more prominent in EGFR TKIresistant cells than parent cells [fig_ref] Figure 2: ERK inhibitor ASN007 induces cell death in EGFR TKIresistant NSCLC cell lines [/fig_ref].
## Asn007 induces cell cycle arrest through fra1 regulation
To identify the underlying mechanism by which ASN007 induces apoptosis in the EGFR TKI-resistant cells, we examined ERK substrate FRA1 and p90RSK. The treatment of ASN007 resulted in decreased FRA1 phosphorylation and protein expression in a time-dependent manner in all EGFR TKI-resistant cells , suggesting that FRA1 is more prominent mediator for ERK signaling than p90RSK in our model system. In previous reports, FRA1 directly regulated the expression of cell cycle-related protein, thereby promoting cell proliferation [bib_ref] Expression and function of FRA1 protein in tumors, Jiang [/bib_ref] [bib_ref] An integrative approach unveils FOSL1 as an oncogene vulnerability in KRAS-driven lung..., Vallejo [/bib_ref]. In line with FRA1 inhibition, ASN007 decreased expression of cell-cycle-related proteins such as FoxM1, Aurora A, PLK1, Cyclin D1, and Cyclin B1 in all EGFR TKI-resistant cells . The effects of c Assessment of apoptosis on ASN007 (500 nM) treatment. βactin was used as a loading control Phosphorylation of ERK is increased in NSCLC tumor samples at disease progression after EGFR TKI treatment. a Representative images of p-ERK1/2 staining from paired tumor samples treated with erlotinib, gefitinib, and afatinib. Top, tumor biopsy before EGFR TKI treatment. Bottom, tumor biopsy after EGFR TKI treatment at disease progression. Scale bar: 50 μm b Quantification of p-ERK1/2 staining score in the firstgeneration EGFR TKI (erlotinib and gefitinib) group (n = 18). c Quantification of p-ERK1/2 staining score in the secondgeneration EGFR TKI (afatinib) group (n = 16). d Correlation between the fold increase in p-ERK staining and treatment period. *, P < 0.05; **, P < 0.01 ASN007 on the cell cycle of EGFR TKI-resistant cells were analyzed using flow cytometry. The cell cycle was arrested at the G 0 /G 1 phase in PC9/ER, PC9/OR, and H1975/OR cells following treatment with 500 nM ASN007 for 24 h .
## Asn007 can effectively overcome erk-driven acquired resistance to egfr tki
Based on these findings, we tested whether ASN007 could overcome the acquired resistance to EGFR TKIs. Combination treatment with EGFR TKIs and ASN007 was more effective than either single agent alone with regard to short-term cell viability in PC9/ER, PC9/OR, and H1975/OR cells . Furthermore, ASN007 alone as well as cotreatment with EGFR TKI reduced the cell viability for long-term (10-14 days) incubation . The ASN007 alone or combination with EGFR TKI markedly inhibited the phosphorylation of FRA1 and induced PARP cleavage compared with EGFR TKI alone .
## Combination treatment with erlotinib and asn007 effectively inhibits the growth of pc9/er xenografts in vivo
To determine whether ASN007 could overcome EGFR TKI resistance in vivo, we used PC9/ER xenograft because the majority of patients in this study were treated with firstgeneration EGFR TKIs. Xenograft tumors induced by PC9/ ER cells continued to grow in vivo with or without erlotinib treatment for 28 days, indicating erlotinib resistance. ASN007 alone significantly decreased tumor growth and the combination of ASN007 with erlotinib completely inhibited tumor a Analysis of FRA1 and p90RSK, ERK-downstream signaling, and activation upon ASN007 (500 nM) treatment. b Analysis of mitosis-related protein expression after ASN007 (500 nM) treatment. β-actin was used as a loading control. c Cell cycle analysis after 24 h of ASN007 (500 nM) treatment (n = 3) growth in PC9/ER xenografts . Consistent with growth inhibition, ASN007 treatment in combined with erlotinib showed further inhibition of FRA1 phosphorylation and FoxM1 expression compared to ASN007 alone . In addition, cell proliferation was synergistically inhibited by combination therapy with erlotinib and ASN007, as assessed by Ki-67 expression . These findings confirm the in vitro results and support the role of ERK signaling in promoting acquired resistance to EGFR TKI.
# Discussion
Several studies have shown that NSCLC cells treated with EGFR TKIs adopt multiple mechanisms to reactivate ERK signaling. The upregulation of phosphorylated ERK following EGFR TKI resistance was a ubiquitous event in EGFR TKI-resistant NSCLC tissues and cell lines. Our study suggested that targeting ERK is an effective strategy to overcome acquired resistance to EGFR TKIs in NSCLC. Furthermore, co-targeting EGFR and ERK will be an even more effective combination strategy.
ERK (ERK1/2) controls several downstream cytoplasmic and nuclear targets by phosphorylating and regulating cell cycle and negative feedback mechanisms [bib_ref] Targeting ERK1/2 protein-serine/threonine kinases in human cancers, Roskoski [/bib_ref]. Although specific ERK alterations have not been identified as actionable mechanisms of acquired resistance to TKIs, alterations in upstream of ERK are common in acquired resistance to various TKIs including EGFR TKI [bib_ref] Overcoming drug resistance to receptor tyrosine kinase inhibitors: Learning from lung cancer, Kuwano [/bib_ref] [bib_ref] Heterogeneous mechanisms of primary and acquired resistance to third-generation EGFR inhibitors, Ortiz-Cuaran [/bib_ref]. Previous studies reported that acquired resistance to EGFR TKIs converged on the activation of the MAPK pathway, especially ERK, albeit through different mechanisms such as MET amplification and RAS alteration [bib_ref] Acquired resistance to AZD9291 as an upfront treatment is dependent on ERK..., Ku [/bib_ref] [bib_ref] Adaptive and acquired resistance to EGFR inhibitors converge on the MAPK pathway, Ma [/bib_ref] [bib_ref] Met gene amplification and protein hyperactivation is a mechanism of resistance to..., Shi [/bib_ref]. Recently, reactivation of ERK signaling through chemokine receptor CXCR7 has been ERK-dependent resistance to EGFR TKI is overcome by combination treatment with ASN007. a Cell viability after EGFR TKI alone, ASN007 alone, or combination treatment for 72 h. Data are presented as mean ± SEM (n = 6). b Colony formation after EGFR TKI and ASN007 combination treatment for 10-14 days. c Western blot analysis showing FRA1, p90RSK, and PARP levels after 24-h treatment with EGFR TKI alone, ASN007 alone, and a combination of EGFR TKI and ASN007. β-actin was used as a loading control identified as a resistance mechanism to EGFR TKI in patients with NSCLC [bib_ref] CXCR7 reactivates ERK signaling to promote resistance to EGFR kinase inhibitors in..., Becker [/bib_ref]. Therefore, targeting ERK would be an attractive strategy for the treatment of a variety of tumor types harboring acquired resistance to TKIs.
Several ERK inhibitors including ASN007, ulixertinib, and LY3214996 are being developed in clinical trials as a treatment for advanced solid tumors with RAS-RAF-MAPK pathway alterations [bib_ref] Targeting ERK1/2 protein-serine/threonine kinases in human cancers, Roskoski [/bib_ref]. Some have demonstrated preliminary antitumor activity in preclinical and clinical trials. However, the most common or dose-limiting toxicities observed to date with ERK inhibitors include diarrhea, nausea, fatigue, and rash. Given that ASN007 is a highly potent and selective ERK1/2 inhibitor in the nanomolar range with a long target residence time, it can be hypothesized that toxicities related to AN007 might be reduced compared to other ERK inhibitors.
In our study, ASN007 shows promising antitumor activity as a single agent and in combination with EGFR TKI. ASN007 alone did not cause any dose-limiting toxicities including loss of body weight or skin rash in a xenograft model. Although mice treated with a combination of ASN007 and erlotinib showed body weight loss 1 week after treatment, this loss was recovered after 2 weeks of treatment. These observations suggest that our proposed combination of ASN007 plus EGFR TKI was well tolerated in a mouse xenograft model. Intriguingly, we found that decreased FRA1 expression is the main mechanism of ERK inhibition by ASN007 in EGFR TKI-resistant NSCLC. In contrast, previous studies using other ERK inhibitors such as ulixertinib and LY3214996 demonstrated that the main downstream target of ERK in KRAS-driven tumors is RSK, not FRA1 [bib_ref] Concurrent HER or PI3K inhibition potentiates the antitumor effect of the ERK..., Jiang [/bib_ref] [bib_ref] ERK inhibitor LY3214996 targets ERK pathway-driven cancers: a therapeutic approach toward precision..., Bhagwat [/bib_ref]. Although phosphorylation of RSK was also inhibited by ASN007 treatment, FRA1 protein expression is more abundant than that of RSK1 in our resistant cells. These results suggest that ERK inhibitor could exert antitumor effects via different mechanisms according to tumor context. FRA1 is a member of the FOS protein family and can form an AP-1 transcription factor. FRA1 is mainly regulated by post-translational phosphorylation by a mitogen-activated protein kinase (MAPK) signaling pathway, especially ERK. Because phosphorylation of FRA1 prevents degradation by ubiquitin-independent proteasome, ERK activation is required for FRA1 accumulation. Many studies have shown that FRA1 is overexpressed in many tumors such as lung cancer, breast cancer, colorectal cancer and other tumors. The abnormal expression of FRA1 in tumor has important roles during tumor progression, promoting cell proliferation and invasion, inhibiting apoptosis, and enhancing tumor angiogenesis and heterogeneity [bib_ref] Expression and function of FRA1 protein in tumors, Jiang [/bib_ref] [bib_ref] FRA-1 as a driver of tumour heterogeneity: a nexus between oncogenes and..., Dhillon [/bib_ref] [bib_ref] The Fra-1-miR-134-SDS22 feedback loop amplifies ERK/JNK signaling and reduces chemosensitivity in ovarian..., Wu [/bib_ref]. Previous studies reported that FRA1 promotes KRAS-induced lung cancer progression and metastasis [bib_ref] An integrative approach unveils FOSL1 as an oncogene vulnerability in KRAS-driven lung..., Vallejo [/bib_ref] [bib_ref] FOSL1 promotes kras-induced lung cancer through amphiregulin and cell survival gene regulation, Elangovan [/bib_ref] [bib_ref] Inhibitor of differentiation-1 sustains mutant KRAS-driven progression, maintenance, and metastasis of lung..., Roman [/bib_ref]. In addition, FRA1 contributed to oncogenic KRAS-driven PD-L1 expression in high risk, premalignant human bronchial epithelial cells, suggesting that FRA1 may promote cancer progression by facilitating immune evasion [bib_ref] 2020) FRA1 contributes to MEK-ERK pathway-dependent PD-L1 upregulation by KRAS mutation in..., Lee [/bib_ref]. Altogether, FRA1 may be a prognostic marker and The combination of ASN007 with erlotinib effectively inhibits the growth of erlotinib-resistant xenografts. a Tumor growth of PC9/ER xenograft treated with vehicle, erlotinib (25 mg/kg/d), ASN007 (50 mg/kg/d), and a combination of erlotinib and ASN007 by oral gavage for 5 days each week. b Western blot analysis of FRA1 and FoxM1 in tumor samples treated as indicated. β-actin was used as a loading control. c Immunohistochemical analysis using Ki-67. Scale bar: 50 μm. *, P < 0.05; **, P < 0.01 potential target for lung cancer with oncogenic mutations or drug resistance.
Our study has several limitations. First, the EGFR TKI resistance models in this study may not represent the variety of clinical situations in patients with EGFR TKI resistance observed in clinic. Although in vitro models we developed in this study have different genomic alterations in upstream of ERK as resistance mechanism, it converged ERK activation regardless of EGFR TKIs. Second, the regulation of acquired resistance by ERK has not been fully elucidated. Further mechanistic studies are needed to investigate how increased ERK signaling exclusively activates the development of drug resistance.
In summary, our findings suggest that concomitant EGFR and ERK blockade is a promising strategy to overcome acquired resistance in EGFR-mutated NSCLC regardless of whether the acquired resistance arises from first-or thirdgeneration EGFR TKIs. Further research is needed to determine whether these combinations can also prevent or delay the development of acquired resistance.
Data availability The data and material generated and analyzed during the present study are available from the corresponding author on reasonable request.
# Declarations
Conflict of interest The authors declare no conflicts of interest. Consent to participate Written informed consent was provided by patients; in some cases, a waiver of consent was obtained.
Consent for publication Written informed consent was provided by patients; in some cases, a waiver of consent was obtained.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
[fig] Figure 2: ERK inhibitor ASN007 induces cell death in EGFR TKIresistant NSCLC cell lines. a Cell viability of erlotinib-resistant (PC9/ER) and osimertinibresistant (PC9/OR and H1975/ OR) cells after EGFR TKI treatment. b Basal protein expression levels in EGFR TKIresistant NSCLC cell lines. [/fig]
|
Novel LMX1B mutation in familial nail-patella syndrome with variable expression of open angle glaucoma
Purpose: To describe the genetic and clinical findings in a large Spanish pedigree with nail-patella syndrome (NPS) and to investigate the expressivity of open angle glaucoma (OAG) in the family members. Methods: All individuals underwent a complete ophthalmologic examination, including optical coherence tomography (OCT) of the optic disc and peripapillary region and ultrasound pachymetry. Screening for mutations in the LMX1B gene was performed by denaturing gradient gel electrophoresis and direct genomic sequencing analysis. Results: Ten family members had NPS, seven with varying degrees of ocular hypertension (OHT). Only one of these had advanced OAG. The others showed high pachymetry values and OCT retinal nerve fiber layer (RNFL) thickness above the normal values. Screening for mutations in the exonic and flanking sequences of the LMX1B gene showed a deletion of one G (289delG) within the coding sequence of exon 3 at codon 97, resulting in a frame shift that creates a premature stop at codon 105 (E97fsX105), predicting a truncated protein. This mutation was present in all NPS patients and absent in the unaffected family members. Conclusions: A novel mutation in the homeobox transcription factor LMX1B causes NPS in a family with variable expressivity of the syndrome, including OAG. The pathogenic mechanism resulting from the mutation is presumably haploinsufficiency rather than a dominant negative effect, which would explain the clinical variability in this family. All NPS OHT patients had considerably thick corneas and RNFL.
Nail-patella syndrome (NPS; OMIM 161200), also known as onychoosteodysplasia, Turner-Kieser syndrome, or Fong disease is a rare heterogeneous, autosomal-dominant pleiotropic disorder characterized by skeletal abnormalities such as nail dysplasia and hypoplastic patellae, renal disease, and open-angle glaucoma (OAG) [bib_ref] Cosegregation of open-angle glaucoma and the nail-patella syndrome, Lichter [/bib_ref] [bib_ref] Nail-patella syndrome and its association with glaucoma: a review of eight families, Mimiwati [/bib_ref] [bib_ref] An orthopaedic scoring system for nail-patella syndrome and application to a kindred..., Farley [/bib_ref]. More than 60 mutations in the LMX1B gene have been associated with NPS, which has a high intra-and interfamily degree of heterogeneity [bib_ref] Nail-patella syndrome: identification of mutations in the LMX1B gene in Dutch families, Knoers [/bib_ref] [bib_ref] Nail patella syndrome: a review of the phenotype aided by developmental biology, Sweeney [/bib_ref].
The identification of the causative gene for this syndrome, the LIM-homeodomain transcription factor LMX1B, led to investigation of the phenotype and molecular pathogenesis in different target organs in NPS patients [bib_ref] Fine mapping of the nail-patella syndrome locus at 9q34, Mcintosh [/bib_ref] [bib_ref] Mutations in LMX1B cause abnormal skeletal patterning and renal dysplasia in nail..., Dreyer [/bib_ref] [bib_ref] Limb and kidney defects in Lmx1b mutant mice suggest an involvement of..., Chen [/bib_ref]. The LMXhomeodomain proteins are a family of transcription factors frequently involved in pattern formation during embryonic development [bib_ref] Induction of the LIM homeobox gene Lmx1 by WNT7a establishes dorsoventral pattern..., Riddle [/bib_ref] [bib_ref] Dorsal cell fate specified by chick Lmx1 during vertebrate limb development, Vogel [/bib_ref]. The LMX1B gene encodes a transcription factor protein that contains two zinc-binding LIM domains (A and B) at the NH 2 -terminus and one homeodomain in the middle, which is responsible for the binding to DNA [bib_ref] The LIM domain: a new structural motif found in zinc-finger-like proteins, Sanchez-Garcia [/bib_ref] [bib_ref] Molecular dissection of a LIM domain, Schmeichel [/bib_ref] [bib_ref] The LIM/double zinc-finger motif functions as a protein dimerization domain, Feuerstein [/bib_ref]. Most of the mutations reported lead to the absence or inactivation of the homeodomain, resulting in a protein unable to recognize its target genes [bib_ref] Mutation analysis of LMX1B gene in nail-patella syndrome patients, Mcintosh [/bib_ref] [bib_ref] Restricted distribution of loss-of-function mutations within the LMX1B genes of nailpatella syndrome..., Clough [/bib_ref].
The LMX1B transcription factor protein is involved in normal patterning of the dorsoventral axis of the embryo during development [bib_ref] Induction of the LIM homeobox gene Lmx1 by WNT7a establishes dorsoventral pattern..., Riddle [/bib_ref] [bib_ref] Dorsal cell fate specified by chick Lmx1 during vertebrate limb development, Vogel [/bib_ref] and early morphogenesis of the glomerular basement membrane [bib_ref] Regulation of glomerular basement membrane collagen expression by LMX1B contributes to renal..., Morello [/bib_ref]. Molecular studies and ge-netic immunochemical experiments carried out in the lmx1b -/mouse model for NPS showed the involvement of lmx1b in the transcription regulation of different alpha chains of type IV collagen in renal podocytes [bib_ref] Transcriptional induction of slit diaphragm genes by Lmx1b is required in podocyte..., Miner [/bib_ref] [bib_ref] The LIM-homeodomain transcription factor Lmx1b plays a crucial role in podocytes, Rohr [/bib_ref]. The defect in collagen fibrillogenesis is also seen in the cornea of lmx1b -/mutant mice, thus supporting the role of LMX1B in collagen regulation. These collagen defects in the eye could be responsible for the presence of glaucoma in NPS patients [bib_ref] LMX1B, a LIM homeodomain class transcription factor, is necessary for normal development..., Pressman [/bib_ref].
NPS is caused by a mutation in the LMX1B gene that presumably leads to loss of function of the encoding transcription factor, and the clinical phenotype is related to the pleiotropic effect of LMX1B during development [bib_ref] Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail-patella syndrome, Vollrath [/bib_ref] [bib_ref] Functional characterization of LMX1B mutations associated with nail-patella syndrome, Sato [/bib_ref]. It has been suggested that as the LMX1B gene encodes a LIM homeodomain transcription factor, which contributes to transcriptional regulation of glomerular membrane collagen expression by podocytes, both renal and ocular malformations could result from the same alteration during embryogenesis.
Many articles have cited the presence of OAG or ocular hypertension (OHT) as one of the features of NPS but the clinical details of this type of glaucoma have never been fully described. Recently, there has been an important advance in glaucoma diagnostic tools with the appearance of new imaging devices of the optic nerve head (ONH) and retinal nerve fiber layer (RNFL), such as optical coherence tomography (OCT) [bib_ref] Evaluation of retinal nerve fiber layer, optic nerve head, and macular thickness..., Medeiros [/bib_ref]. Moreover, the role of pachymetry in the differential diagnosis of OHT and glaucoma is progressively gaining more importance [bib_ref] Corneal thickness in glaucoma screening, diagnosis, and management, Brandt [/bib_ref] prior to initiating expensive hypotensive treatments that can cause discomfort and side effects in these patients. Our study had two aims: (1) to screen a seven-generation NPS Spanish family for mutations in the LMX1B gene; and (2) to characterize the clinical phenotype of their OAG or OHT by performing new structural optic disc examinations such as OCT.
# Methods
Patients: Prior to their inclusion in this study, all participants were informed of its objectives. A total of 18 family members (12 women and 6 men) ranging in age from 7 to 90 years were analyzed. All gave their informed consent to participate in the study, which adhered to the tenets of the Declaration of Helsinki. Affected and unaffected members of a NPS Spanish family underwent a complete ophthalmologic examination and DNA genetic analysis. During the physical examination, some patients mentioned they were receiving topical glaucoma medications. After the initial examination, all patients were scheduled for several follow-up visits over three years. One patient, (IV-1), died a few months after the initial visit before undergoing other tests, such as pachymetry or OCT.
Molecular genetics: Blood samples were drawn for DNA analysis. Genomic DNA was prepared from peripheral blood lymphocytes using QIAmp DNA blood mini kit (Qiagen, Valencia, CA). The coding region of the LMX1B (NM_00231) gene was amplified using primers located in the flanking intron region. Further amplification of exon 3 of LMX1B was carried out using 5'-CTG GGA GGG ACT TCT GAG CA-3' as the forward primer and 5'-CTC CAG GAC ACC CCA GCA AC-3' as the reverse primer. A "CG clamp", consisting of a sequence of 40 CG nucleotides, was included in the 5' sequence of the forward primer, 5'-CCT CCA GGA CAC CCC AGG AA-3', to provide better resolution in the denaturing gradient gel electrophoresis (DGGE) analysis [bib_ref] Attachment of a 40-base-pair G + C-rich sequence (GC-clamp) to genomic DNA..., Sheffield [/bib_ref]. Polymerase chain reaction (PCR) was performed in a 50 µl volume of buffer (20 mM Tris-HCl, pH 8.55; 16 mM (NH) 2 SO 4 ; 1.5 mM MgCl 2 ; 150 µg/ml BSA) containing 200-500 ng of human genomic DNA, 25 picomols of each primer, 10 nanomols of each deoxyribonucleoside triphosphate, and 1.5 units of Taq polymerase (Ecotaq). Incubation was performed for 40 cycles consisting of 1 min at 94 °C, 1 min at 50 °C, and 1 min at 72 °C; this was followed by 1 min at 94 °C and 5 min at 72 °C. Electrophoresis of 8 µl of final PCR reaction volume was performed on 1.5% agarose gel to test the amplification reaction. For DNA sequencing, PCR products were purified using Qiaquick columns (Qiagen). DNA sequencing was carried out using the OpenGene automated DNA sequencing system from Visible Genetics and Thermo Sequenase Cy5.5 Dye Terminator Cycle Sequencing kit (Amersham Pharmacia Biotech, Barcelona, Spain). The sequencing primers were the same as those used for the PCR reaction.
Eye examination: The personal and family medical history of each patient was critical for determining which individual was affected and for establishing the inheritance pattern. Details about other systemic NPS anomalies were obtained from the each participant's medical chart. A comparison of the severity of systemic manifestations between the family members was performed to establish a spectrum of systemic involvement.
Best corrected visual acuity (BCVA) was tested with Snellen charts, and ocular refraction was evaluated with an autorefractometer. Biomicroscopy of the anterior segment was performed. Tonometry was done with a Goldmann applanation tonometer, using the mean of three consecutive tonometry readings for each eye. Gonioscopy with a Goldmann threemirror lens and funduscopy for the observation of the ONH and peripapillary region under direct examination with a 78 D Volkmann lens completed the slit-lamp exam.
Ancillary examinations: All cases underwent computerized perimetry with two types of perimeters to confirm the accuracy of the data. We used a 750-Humphrey field analyzer II following a SITA Standard strategy and an Octopus 101 perimeter, TOP strategy. We used a size III stimulus and a white-on-white test for both devices.
Monoscopic ONH and peripapillary RNFL digital pictures were taken with a Topcon nonmydriatic retinal camera (Topcon model TRC-NW6S, Topcon, Tokyo, Japan).
Ultrasound pachymetry was performed under topical anesthesia with an Ocuscan RxP pachymeter (Alcon Laboratories Inc., Irvine, CA) that provides six central readings and calculates the mean value. The intraocular pressure (IOP) is then manually introduced in each case and the machine automatically provides a new IOP value corrected for the pachymetry measurements (Herndon Formula).
Optical coherence tomography was performed with OCT version 3 (Stratus OCT, Carl Zeiss Meditec Inc., Dublin, CA), which is a high-resolution cross-sectional imaging technique that allows in vivo measurements of the RNFL (described further in reference [bib_ref] Optical coherence tomography, Huang [/bib_ref]. All subjects received pupillary dilatation with 0.5% tropicamide.
Patients underwent three different protocols: fast RNFL thickness, fast ONH, and fast macular thickness measurements. The studied parameters in each protocol (average thickness, superior average thickness, inferior average thickness, vertically integrated rim area, horizontally integrated rim width, cup/disc area ratio, inferior outer macula, and temporal outer macula) were the most relevant for the glaucoma diagnosis according to the study by Medeiros; as all had receiver operating characteristic (ROC) curves above 0.75 except the disc area that was included in our study for demographic purposes [bib_ref] Evaluation of retinal nerve fiber layer, optic nerve head, and macular thickness..., Medeiros [/bib_ref].
In all protocols only images with a quality above 7 were accepted. Exams were repeated until this level was reached.
# Results
# Genetic analysis:
We screened for mutations in all of the exonic and flanking sequences of LMX1B: (NM_002307) in the index case of an NPS family. A DGGE alteration pattern was detected in exon 3 of this index patient but absent in 56 unrelated controls. Direct genomic sequencing revealed the deletion of one G (289delG) within the coding sequence of exon 3 at codon 97 of LMX1B . This nucleotide deletion changes the open reading frame of the gene, creating a prema- . Genetic results of the NPS family. A: Pedigree of a Spanish family with nail-patella syndrome (NPS). Filled squares and circles denote NPS-affected males and females, respectively. Open squares and circles represent males and females without NPS. A diagonal line through the symbols marks a deceased individual. An arrow indicates the index patient (VI-4). The following symbols were used: wild-type allele (+)2889delG allele (-), and 372A>G allele (x), correspond to a, and allele, respectively. B: Denaturing gradient gel electrophoresis (DGGE) of the PCR fragment of exon 3. The different electrophoretic profiles correspond to carriers of the following: 289delG/WT (IV-1, IV-2, IV-3, V-1, V-2, V-6, and VI-2); WT/372A>G (V-3, VI-1, VI-6, VI-7); 289delG/372A gt G (V-5, V-9, VI-4); WT/ WT (V-4, V-8, VI-3) and 372A>G/372A>G (VI-5). WT=wild type. C: DNA genomic sequencing of exon 3 of LMX1B showing the 289delG mutation (left) and a control DNA (right). ture stop at codon 105 (E97fsX105), presumably giving rise to a truncated protein. Both DGGE analysis and sequencing of exon 3 were carried out in all members of this family. Analysis of the family revealed that several members carried a novel nucleotide change 372 A>G that generates the silent mutation Glu124Glu (data not shown). Because mutations 289delG and 372>G in LMXB1 segregate separately, as deduced by DGGE and sequencing analysis, both mutations would be in separate chromosomes (different alleles). It has been hypothesized that the severity of the NPS may also be related to the second copy of LMXB1 gene [bib_ref] An orthopaedic scoring system for nail-patella syndrome and application to a kindred..., Farley [/bib_ref]. In this large family we cosegregated both alleles (containing the NPS and the silent mutation), but no evidence supporting the hypothesis of a second allele interaction could be found.
Ten family members had NPS with varying degrees of systemic involvement [fig_ref] TABLE 1: SYSTEMIC PHENOTYPE OF NPS FAMILY [/fig_ref] , although one NPS member died (IV-1) during follow-up. The inheritance pattern was autosomal dominant with complete penetrance but variable expressivity . Onychodysplasia [fig_ref] Figure 2: Gonioscopy of NPS patients [/fig_ref] and pa-tellar hypoplasia were the only signs present in all the patients.
Nephropathy (glomerulonephritis) and colon disease (spasmodic colitis) appeared in only a few members. Different skeletal abnormalities were observed in some members of the family: Patients V-5 and VI-4 showed hip subluxation, and patient VI-4 also had generalized joint hyperextensibility, particularly in the fingers. Some patients described generalized muscular pains (V-5, VI-2, VI-4) and chronic fatigue (V-5). The severity of the systemic involvement of NPS was unrelated to patient age, with the index patient being not only the youngest but also one of the most seriously affected.
Ophthalmologic features: Seven of the NPS patients reported having OAG were treated by their ophthalmologists with one or two topical hypotensive drugs. Five of these complained of ocular side effects such as conjunctival hyperemia, itching, iris hyperpigmentation, and eyelash growth. These are well known common side effects of such hypotensive agents [fig_ref] TABLE 2: OPTHTHALMOLOGIC FEATURES OF NPS PATIENTS [/fig_ref].
[formula] ------- --- --- --------- ----------- ------------- ----------- --------- --- IV-1 86 F + + + + - + IV-2 72 M + + + - - + IV-3 67 F + + + - + + V-1 60 M + + + - + + V-2 56 F + + + - - - V-5 46 F + + + - - + V-6 41 F + + + + - + V-9 29 M + + - - - + VI-2 32 F + + + - - - VI-4 28 F + + + + + - [/formula]
Systemic anomalies of all nail-patella syndrome (NPS) patients evaluated in this study. The following terms were used: F refers to female, M refers to male, OAG is open angle glaucoma, and OHT is ocular hypertension.
BCVA was severely diminished in the eldest patient due to diabetic retinopathy with macular edema (IV-1). Two patients with cataracts (IV-2, IV-3) underwent uneventful bilateral phacoemulsification, and their BCVA was unremarkable in the postoperative period. A 48-year-old patient (V-5) presented with mild lens opacities that caused a certain degree of visual loss, suggesting the presence of congenital cataracts. The refractive status of all patients was within the normal range (data not shown).
Subjects receiving topical hypotensive medications were asked to stop their medications for a wash-out period of three to four weeks in order to establish a baseline intraocular pressure. Gonioscopy revealed the presence of multiple iris processes [fig_ref] Figure 2: Gonioscopy of NPS patients [/fig_ref] in the majority of the OAG patients and a highly pigmented trabeculum was also quite common, markedly so in one case (V-9; [fig_ref] Figure 2: Gonioscopy of NPS patients [/fig_ref]. No dysplastic changes or presence of abnormal vessels were noticed. All had wide, open anterior chambers and angles measured more than 45°, except in one case of 30°in both eyes (V-5). The measurement numbers in red show thickness measurements above the normal range; the rest of the boxes show normal measurements. One patient (V-1) presented with vertical saucerization in his right ONH and bilateral optic disc asymmetry, defined by a difference greater than 0.2 with left eye cupping. The eldest patient (IV-1) presented with a cup/disc ratio of 0.9, bayoneting of vessels and severe pallor, suggesting advanced glaucoma neuropathy.
[formula] - --- ------ ------ --- ----- ----- ----- ----- ----- ----- --- --- IV-2ODLMX1B VA VA IOP IOP PACH PACH C IOP C IOP C/D C/D Case haplotype OD OS OD OS OD OS OD OS OD OS ---- --------- --- --- --- --- ---- ---- ----- ----- --- ---- IV-1 -/+ LP [/formula]
Ultrasound pachymetry displayed high readings in all NPS patients in contrast with other non-NPS family members who had mean pachymetry readings of 530±3 µm.
Computerized perimetry was unremarkable in seven patients. Two (IV-2, IV-3) had diffuse decreased sensitivity due to their senile cataracts and presented with normal visual field testings after their cataract extraction. Patient V-5 had an initial inferior arcuate scotomata in both eyes, more marked in her right eye and mild diffuse loss due to congenital lens opacity (Octopus data in decibels (dB): MD 3.7; LV 9.9, OD/MD 2.2, LV 6.4 OS, and Humphrey perimeter data in dB: MD -5.14, DSM 3.57, OD/MD -2.40, DSM 2.09 OS). Patient IV-1 could not perform this test due to her severe low vision.
Stratus optical coherence tomography data: The Fast RNFL protocol showed extraordinary high thickness measurements through consecutive OCT tests , showing that the peripapillary nerve fiber layer was thicker in these patients than in the normal population (data from 350 controls col- lected on the hard disk of the OCT machine). Five NPS patients (four with OHT and one normotensive) presented with mean thickness measurements in the "white zone" of the OCT plot [fig_ref] Figure 3: Optical coherence tomography of patients IV-2 and V-1 [/fig_ref].
The Fast ONH protocol provided measurements of optic disc and cup diameters, volumes, and ratios that were also unremarkable except in patient V-I [fig_ref] Figure 3: Optical coherence tomography of patients IV-2 and V-1 [/fig_ref]. This patient presented with a pathologic cup/disc (C/D) area ratio in his OD (0.8) and bilateral asymmetry above 0.2 (C/D ratio 0.6 OS).
The Fast macular thickness protocol provided completely normal results for all the patients.
Glaucoma therapeutic approach: Patient IV-I had been previously diagnosed with severe OAG based on tonometry and optic disc aspect and was under treatment with two drugs until she died.
Patients IV-2 and IV-3 were diagnosed with OHT or preperimetric OAG, and their topical hypotensive drugs were stopped. They presented normal perimetry and OCT at follow-up examinations after bilateral phacoemulsification.
Patient V-1 was diagnosed with glaucoma suspect upon optic nerve asymmetry between both eyes, confirmed by the C/D area ratio provided by OCT. He is being treated with one hypotensive drug (prostaglandin analog nightly).
Patient V-2 and her two daughters (VI-2, VI-4) were considered as non-OHT NPS patients.
Patient V-5 was deemed to have glaucoma based on the IOP readings and initial arcuate perimetry lesions, although the optic nerves were not apparently affected according to their aspect and the OCT. She is currently receiving hypotensive treatment with a prostaglandin analog and, as she is developing cataracts, will be scheduled for phacoemulsification.
Patients V-6 and V-9 were diagnosed with OHT because their IOPs were above 20 mmHg but less than 30 mmHg. Their ophthalmologic explorations were otherwise unremarkable. They currently remain without treatment.
# Discussion
We screened an index case from a large Spanish family with NPS for mutations in the LMX1B gene and found a novel nonsense mutation, 289delG, causing a premature stop at codon 105 (E97fsX105). The mutation is located in exon 3 of LMX1B, which encodes the COOH-terminal region of the LIM-A domain. Thus, in the putative-encoded mutant protein, the lack of the LIM-B domain and the homeodomain (HD) probably impede the DNA binding capacity of the protein.
To date, more than 60 point mutations, small deletions, or insertions have been reported, clustered in the LIM and HD domains of the LMX1B gene [bib_ref] Mutation analysis of LMX1B gene in nail-patella syndrome patients, Mcintosh [/bib_ref] [bib_ref] Restricted distribution of loss-of-function mutations within the LMX1B genes of nailpatella syndrome..., Clough [/bib_ref] [bib_ref] Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail-patella syndrome, Vollrath [/bib_ref] [bib_ref] Functional characterization of LMX1B mutations associated with nail-patella syndrome, Sato [/bib_ref]. However, no mutations in NPS patients have been detected in the COOH-terminal third of the coding sequence of the LMX1B gene, suggesting that mutations in this region are not inactivating [bib_ref] Restricted distribution of loss-of-function mutations within the LMX1B genes of nailpatella syndrome..., Clough [/bib_ref]. Furthermore, transfection studies performed with NPS mutants showed no significant dominant-negative effect on the function of the wild-type LMX1B [bib_ref] Functional characterization of LMX1B mutations associated with nail-patella syndrome, Sato [/bib_ref] [bib_ref] LMX1B transactivation and expression in nail-patella syndrome, Dreyer [/bib_ref]. These findings support haploinsufficiency as the mechanism underlying pathogenesis of NPS [bib_ref] Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail-patella syndrome, Vollrath [/bib_ref] [bib_ref] Functional characterization of LMX1B mutations associated with nail-patella syndrome, Sato [/bib_ref] [bib_ref] Genotype-phenotype studies in nail-patella syndrome show that LMX1B mutation location is involved..., Bongers [/bib_ref].
Mutations in LMX1B associated with NPS result in skeletal defects. All family members in this study who carried the E97fsX105 mutation showed nail dysplasia and hypoplastic patellae [fig_ref] TABLE 1: SYSTEMIC PHENOTYPE OF NPS FAMILY [/fig_ref] , and most of them also had other skeletal abnormalities. These common features were present with almost complete penetrance in the family. Some typical characteristics were present at an early age, with members of the family being able to predict who were carriers of the disease. Indeed, their predictions agreed with our results concerning the carriers of the mutation in the LMX1B gene. However, renal, gastrointestinal, or ophthalmic involvement was more variable within the family.
The expression of LMX1B in the kidney occurs exclusively in podocytes [bib_ref] The LIM-homeodomain transcription factor Lmx1b plays a crucial role in podocytes, Rohr [/bib_ref]. Studies carried out in lmx1b knockout mice that mimic NPS showed an important impairment of the glomerular basement membrane with expression of the α3 and α4 chains of collagen IV and severe reduction of podocin [bib_ref] The LIM-homeodomain transcription factor Lmx1b plays a crucial role in podocytes, Rohr [/bib_ref] [bib_ref] LMX1B, a LIM homeodomain class transcription factor, is necessary for normal development..., Pressman [/bib_ref]. Importantly, lmx1b is also necessary for normal development of multiple tissues in the anterior segment of the murine eye [bib_ref] LMX1B, a LIM homeodomain class transcription factor, is necessary for normal development..., Pressman [/bib_ref] Abnormalities in collagen fibrils in the corneal stroma of lmx1b knockout mice have been observed. The abnormal expression of collagen in homozygous lmx1b knockout mice may explain the split in the glomerular basement membrane in the kidney or the eye abnormalities [bib_ref] The LIM-homeodomain transcription factor Lmx1b plays a crucial role in podocytes, Rohr [/bib_ref] [bib_ref] LMX1B, a LIM homeodomain class transcription factor, is necessary for normal development..., Pressman [/bib_ref] possibly related with glaucoma. Surprisingly, lmx1b +/mice do not show an abnormal phenotype, indicating differences in the lmx1b gene doses compared to humans. Recent immunohistological studies in NPS patients with severe glomerular disease suggest a possible regulation of type III collagen by LMX1B, while the homozygous mutations of LMX1B do not appear to dramatically affect the expression of type IV collagen chains and podocin in NPS patients.
Several well-known NPS studies have reported cosegregation of OAG and NPS as a result of a pleiotropic effect of the LMX1B gene in different families [bib_ref] Cosegregation of open-angle glaucoma and the nail-patella syndrome, Lichter [/bib_ref] [bib_ref] Nail-patella syndrome and its association with glaucoma: a review of eight families, Mimiwati [/bib_ref] [bib_ref] An orthopaedic scoring system for nail-patella syndrome and application to a kindred..., Farley [/bib_ref] [bib_ref] Mutation analysis of LMX1B gene in nail-patella syndrome patients, Mcintosh [/bib_ref]. None of these studies described the glaucomatous features in detail. Some ocular abnormalities have been described in the knockout mouse model [bib_ref] LMX1B, a LIM homeodomain class transcription factor, is necessary for normal development..., Pressman [/bib_ref] such as an abnormal keratocan expression in the cornea, abnormal collagen fibrillogenesis, and a reduction in the depth of the anterior chamber. The LMX1B gene is also expressed in the trabecular meshwork but abnormalities in this tissue were not analyzed due to the short life span of the knockout mouse. In addition, many other ocular anomalies have been found sporadically in NPS patients, including microcornea, sclerocornea, congenital cataracts, iris processes, and pigmentation of the inner margin of the iris (referred to as Lester's sign) [bib_ref] Cosegregation of open-angle glaucoma and the nail-patella syndrome, Lichter [/bib_ref] , but none of these is a specific trait of this syndrome.
Sweeney et al. [bib_ref] Nail patella syndrome: a review of the phenotype aided by developmental biology, Sweeney [/bib_ref] provided data from 123 NPS patients, most of whom were visited at home. IOP was measured using a Tonopen® and patients with IOP readings higher than 21 mmHg were advised to seek an ophthalmologist for further investigation. The researchers reported a prevalence of glaucoma and OHT of 9.6% and 7.2%, respectively. No other additional information of note for a glaucoma diagnosis is available in their study. Bongers et al. [bib_ref] Genotype-phenotype studies in nail-patella syndrome show that LMX1B mutation location is involved..., Bongers [/bib_ref] gave more reliable information. In their study, patients underwent gonioscopy and Humphrey visual field analysis. They found glaucoma and OHT in 35.3% of the patients and reported other ocular anomalies, such as corneal anomalies, iris pigmentation, pigment dispersion syndrome, and congenital cataracts. Lichter et al. [bib_ref] Cosegregation of open-angle glaucoma and the nail-patella syndrome, Lichter [/bib_ref] reported OAG in 13 of 33 families, some of whom had IOP higher than 30 mmHg and required filtration surgery. Recently, a large study by found glaucoma in 11.1% of his NPS population (33% of cases over the age of 40 years) and OHT in 11.1%. They reported an average age at diagnosis of glaucoma in cases with apparently LMX1B causative mutations of 29.7 years, which is substantially below the expected age at diagnosis for most cases of primary OAG. The authors performed a comprehensive glaucoma screening that included IOP, automated perimetry, and stereo disc photographs on all available family members.
New imaging techniques, such as OCT, are becoming useful tools for the structural analysis of the optic disc and peripapillary region, target sites of glaucomatous damage. The most useful protocols are the Fast RNFL thickness and Fast ONH, and several OCT parameters are presently being used to evaluate the presence of glaucomatous optic disc damage [bib_ref] Evaluation of retinal nerve fiber layer, optic nerve head, and macular thickness..., Medeiros [/bib_ref]. This is the first report of OCT data in NPS patients with glaucoma or OHT. Surprisingly, all our study patients presented RNFL thickness measurements above normal. This indicates that there is no loss of peripapillary nerve fiber bundles and thus excludes the presence of glaucomatous optic neuropathy in our study population. These data correlated well with the aspect of the optic disc and the perimetric results, except in patient V-5, who presented arciform scotomata. This patient requires future perimetry testing after phacoemulsification, as the cataract could skew the perimetry results. One patient (V-1) presented with marked optic disc cupping according to the examiner's subjectivity; it was supported by the ONH OCT protocol.
Pachymetry-measured central corneal thickness has a significant effect on the clinical management of patients with glaucoma and glaucoma suspects [bib_ref] Clinical significance of central corneal thickness in the management of glaucoma, Shih [/bib_ref]. In 1970, Ehlers [bib_ref] On corneal thickness and intraocular pressure. II. A clinical study on the..., Ehlers [/bib_ref] published a paper suggesting that corneal thickness, another biomechanical measurement of the eye, also affected IOP. Technology to measure corneal thickness has developed over time, and it was not until the publication of the Ocular Hypertension Treatment Study (OHTS) [bib_ref] Corneal thickness in glaucoma screening, diagnosis, and management, Brandt [/bib_ref] in 2002 that central corneal thickness (CCT) was suggested to be an independent risk factor for progression to initial glaucoma damage among individuals with OHT.
A thin pachymetry reading under 530 µm is considered a risk factor for developing glaucoma in eyes with OHT. Recent general practice guidelines recommend performing routine pachymetry and applying a correction factor to the IOP measurements in order to avoid a confounding effect. An association has even been proposed between CCT and the elasticity of the lamina cribrosa. A very recent study [bib_ref] Central corneal thickness and correlation to optic disc size: a potential link..., Pakravan [/bib_ref] even suggested an inverse correlation between CCT and optic disc area. While thicker corneas may cause examiners to make a slight overestimation of true IOP, their thickness may also in-dicate the presence of a substantially smaller, and thus more robust, ONH. Conversely, individuals with thinner corneas, which may initiate a slight underestimation of true IOP, may also be those with larger and more deformable optic discs. Our study population corroborated this fact: All presented with thick corneas and thick RNFL measurements whereas non-NPS family members had "normal" CCTs. Whether thick CCT belongs to the constellation of NPS traits is still unknown and requires routine pachymetry in all these patients.
Our study population had OHT that fell within "normal" limits of IOP readings after the automated pachymetry corrections performed by the Ocuscan®, as they all had very thick corneas. According to the OHTS [bib_ref] Corneal thickness in glaucoma screening, diagnosis, and management, Brandt [/bib_ref] , they all have a low risk of developing OAG in the future. Moreover, their functional and structural tests were normal, and medical treatment was not required in these cases. It was actually withdrawn in several cases except those with suspected glaucoma. We do not know if any other NPS glaucoma cases described in the literature would have OHT or possible glaucoma after pachymetry IOP correction.
The role that iris processes play in the development of OHT is unknown, as they have classically been considered to be innocuous, unlike the presence of high trabecular pigmentation. Almost all our OHT and non-OHT NPS patients had a myriad of iris processes in their anterior chamber angles. This trait, therefore, seems to accompany the other NPS manifestations, as has been described before, and is independent of IOP.
In the family reported, some patients had OHT and kidney or colon abnormalities. The 30% proportion of the family who had gastrointestinal and renal abnormalities of NPS is similar to previously reported data. The presence of glaucoma or OHT has been estimated to be near 30% in other studies, but in this family 70% of the NPS patients had OHT, indicating that this is a significant feature in the family. OHT was not a constant sign and its severity did not depend on the age of the patient or the degree of systemic disease. The index patient in this family was one of the most affected individuals, with severe skeletal anomalies and nephropathy, though she did not have OHT.
We conclude that NPS shows a high inter-and intrafamily heterogeneity. The pathogenesis of OAG or OHT in these patients is still not well understood, and it shows no correlation with the severity of other systemic anomalies. OHT was not present in unaffected individuals, and it seems to be a genuine pleiotropic effect of the LMX1B mutation in this family.
Caution should be taken when diagnosing the presence of glaucoma in persons with NPS, and it should be corrected by pachymetry. OCT is a useful tool for the diagnosis of glaucomatous optic nerve and RNFL pathology, especially in those cases with unclear perimetric testing. We encourage other investigators to perform pachymetry and OCT in NPS patients in order to corroborate the presence of a higher thickness of the corneal and retinal nerve tissues. Surprisingly both factors could, in a certain way, be considered as "protective" ocular mechanisms against the presence of glaucomatous optic nerve damage. There is still a long way to go in the establishment of a correlation between glaucoma and other ocular manifestations of this syndrome, such as the presence of multiple iris processes. Ultrastructural studies are needed to establish whether both renal and ophthalmic anomalies belong to the same form of inheritable connective tissue disorder.
[fig] Novel: LMX1B mutation in familial nail-patella syndrome with variable expression of open angle glaucoma Elena Millá, 1 Imma Hernan, 2 María José Gamundi, 2 Maria Martínez-Gimeno, 2 Miguel Carballo 2 1 [/fig]
[fig] Figure 2: Gonioscopy of NPS patients. A: Color photograph taken during gonioscopy of multiple iris processes in patient V-6. B: Color photograph of a highly pigmented anterior chamber angle of patient V-9. [/fig]
[fig] Figure 3: Optical coherence tomography of patients IV-2 and V-1. A: Optical coherence tomography (OCT) diagram for patient IV-2 in which abovenormal retinal nerve fiber layer (RNFL) thickness measurements are expressed in white in the pie chart. Green sectors are within the normal range. B: OCT Fast optic nerve head protocol of patient V-1, confirming cup enlargement. [/fig]
[table] TABLE 1: SYSTEMIC PHENOTYPE OF NPS FAMILY [/table]
[table] TABLE 2: OPTHTHALMOLOGIC FEATURES OF NPS PATIENTS [/table]
|
An unusual clinical presentation resembling superior vena cava syndrome post heart surgery
Background: An unusual sequence of post operative events heralded by hemodynamic deterioration followed by dyspnea and rapidly progressive dilatation of superficial neck and facial veins, resembling a superior vena cava syndrome, two days post surgical resection of filamentous aortic valve masses, closure of a patent foramen ovale, and performance of a modified Maze procedure for atrial fibrillation in a patient that presented with transient neurologic findings is presented.Case Presentation: Although both clinical findings and hemodynamic derangements completely resolved following tricuspid valve repair aimed to correct the new onset severe tricuspid regurgitation noted post operatively; a clear mechanism was not readily obvious and diagnostic testing data somewhat conflictive. We present a careful retrospective examination of all clinical data and review possible clinical entities that could have been implicated in this particular case and recognize that transesophageal echocardiographic findings were most useful in identifying the best course of action.Conclusion:After reviewing all clinical data and despite the inconclusive nature of test results; the retrospective examination of transesophageal echocardiographic findings proved to be most useful in identifying the best course of action. We postulate that in our case, resolution of the suspected pulmonary embolism with anticoagulation and reestablishment of a normal right ventricular geometry with tricuspid valve repair worked in unison in restoring normal hemodynamics and resolving both dyspnea and venous dilatation.
# Background
It has been suggested that when patients suffer unexpected findings or develop multiple complications after any intervention, adverse outcomes might be expected if a strategy is not promptly implemented to identify the offending or triggering mechanism(s). In this case we present a patient who underwent surgical resection of filamentous aortic valve masses, closure of a patent foramen ovale, and performance of a modified Maze procedure for atrial fibrillation who then developed a series of unusual post operative events heralded by hemodynamic deterioration, dyspnea and rapidly progressive dilatation of superficial neck and facial veins, resembling a superior vena cava syndrome; two days post surgery. A careful retrospective examination of all clinical data is presented with a detailed review of all possible clinical entities.
## Case report
A 72-year-old woman with chronic atrial fibrillation, temporal arteritis, and avascular necrosis of the right hip requiring hemiarthroplasty that was complicated by deep venous thrombosis in the past, now presents with new onset left upper extremity clumsiness. A head computed tomography (CT) scan and subsequent magnetic resonance imaging study revealed a hemorrhage into the right parietal and occipital regions of the brain. In view of these findings, warfarin therapy was discontinued and as part of the workup a transesophageal echocardiogram (TEE) was performed and revealed a 4 × 6 mm mass on the aortic valve and a patent foramen ovale (PFO) with bidirectional shunting (right to left after the intravenous injection of agitated saline), mild bi-atrial enlargement, mildly dilated but normally functioning right ventricle, normal left ventricular systolic function and no significant valvular regurgitation. The patient was afebrile and blood cultures were negative. The patient was referred to our institution for surgical evaluation given the TEE findings and the patient clinical presentation. Surgical resection of the aortic valve mass and closure of the PFO were recommended.
Upon examination of the aortic valve, a filamentous mass was identified on the leading edge of the noncoronary cusp midway between the ranula and the commisure. The mass was excised with sharp dissection. A second 2 × 3 mm mass was also observed on the underside of the leaflet and was excised in a similar fashion. Lesions were also observed at the commisure and on the left coronary cusp and were also excised. Pathologic examination subsequently revealed that these lesions were small fragments of myxoid tissue with hyalinization. A modified Maze procedure was also performed at the time of surgery and it required mobilization of the pulmonary veins bilaterally with application of radiofrequency lesions just proximal to the entrance of the left and right pulmonary veins. Another radiofrequency lesion was placed at the base of the left atrial appendage and this was then excised and then over sewn. Finally, the right atrium was then opened in a longitudinal fashion, the PFO was identified at the cephalad portion of the fossa ovalis and was then closed in a running fashion.
The patient was weaned from bypass without incident and a TEE performed in the operating room revealed no residual masses on the aortic valve or right to left shunt. How-ever, a mild degree of tricuspid regurgitation was now identified for the first time. Both right ventricular size and systolic function remained normal. The patient was taken to the intensive care unit in stable condition and the immediate post-operative course was unremarkable. There was no need for either vasopressor or ionotropic support and the patient was extubated on the first postoperative day and subsequently transferred to a step down unit.
The patient did well until the night of the second postoperative day when she developed oliguria that progressively worsened followed by tachycardia and hypotension. At this time, a pronounced jugular venous distention was noted and a bedside transthoracic echo was immediately obtained to rule out the presence of pericardial tamponade. No pericardial effusion was noted; however, severe tricuspid regurgitation was now noted with dilatation and dysfunction of the right ventricle. In view of these findings, a spiral chest CT was requested but there was no evidence of pulmonary embolus; however, both superior as well as inferior vena cavae were dilated. The pericardium thickness was reported as normal. Because the clinical deterioration persisted, the patient was taken back to the operating room for mediastinal exploration; only a small amount of pericardial fluid was removed. Although the pericardial sac appeared to be under tension, no clot was found after extensive exploration.
Despite a trial of medical management, the patient's cardiac index remained low and all facial and neck veins became progressively and markedly engorged. A repeated TEE demonstrated marked dilatation of the right-sided chambers with severe reduction in right ventricular systolic function and severe tricuspid regurgitation as a result of an incomplete coaptation of the tricuspid valve leaflets . During this TEE, normal left ventricular contractility was noted with no regional wall motion abnormalities, with the exception of a paradoxical septal wall motion as expected post recent surgery, and normal left ventricular systolic function. In addition, pulsed interrogation of all four pulmonary veins showed normal velocities in order to rule out post-surgical Maze procedure pulmonary vein stenosis. No visible alteration in either the main or secondary pulmonary arteries was seen by TEE or CT. Although a clear unifying mechanism that would encompass the hemodynamic deterioration as well as the disproportionate symptoms of dyspnea with the rapidly progressive dilatation of all superficial neck and facial veins could not be given, it was clearly evident that the incomplete coaptation of the tricuspid valve leaflets, as seen on TEE, would certainly continue to further deteriorate the current clinical condition. Therefore, the patient was taken to the operating room again and upon opening the right atrium, the dilatation of the tricuspid valve annulus was confirmed resulting in failure of coaptation of the otherwise normal appearing tricuspid leaflets. Tricuspid annuloplasty with a 34 mm Cosgrove band was performed with marked improvement in cardiac output not requiring further inotropic support. A postoperative TEE revealed no evidence of residual tricuspid regurgitation and the chest was closed. The remainder of the patient's post-operative course was uncomplicated. Unfractionated intravenous heparin was started once the surgeon felt it was safe to maintain a therapeutic activated partial thromboplastin time. The pronounced engorgement of her neck and face veins resolved and she remained hemodynamically stable. After the 5 th postoperative day, the patient was transferred to a rehabilitation facility. At one-month follow-up, the patient has remained in atrial fibrillation, but is otherwise doing well.
# Conclusion
This case demonstrates an unusual sequence of post operative events that occurred after surgical resection of filamentous aortic valve masses, closure of a patent foramen ovale, and performance of modified Maze procedure for atrial fibrillation in a patient presenting with transient neurologic findings. The inconclusive nature and contradictory results of diagnostic testing made it difficult to accurately identify the responsible mechanism(s) and reminds us that in order to clearly solve any clinical mystery we have to review all possibilities, and despite the unclear nature of objective evidence, whatever remains, though improbable, must in the end make sense and indeed might be the most simple explanation.
Given the most obvious postoperative complication, we ruled out pericardial tamponade and then tried to make sense of the striking dilatation of all superficial neck and facial veins resembling superior vena cava syndrome [bib_ref] Etiologic considerations in superior vena cava syndrome, Parish [/bib_ref] [bib_ref] Superior vena caval obstruction: a 10-year experience, Bell [/bib_ref] [bib_ref] Diagnosis and management of superior vena cava syndrome, Markman [/bib_ref]. The rapidly progressive nature of the venous engorgement was contrary to the known pathophysiology encountered by the clinical entities associated with superior vena cava syndrome that are rather insidious [bib_ref] Diagnosis and management of superior vena cava syndrome, Markman [/bib_ref] [bib_ref] Superior vena cava syndrome: an oncologic complication, Stewart [/bib_ref]. Furthermore, there was no recent or long-standing central catheter or pacer wire in place that would have incidentally caused venous thrombosis [bib_ref] Venous thrombosis and stenosis after implantation of pacemakers and defibrillators, Rozmus [/bib_ref]. Therefore, additional considerations needed to be sought in order to explain this striking finding.
The possibility of embolization from one or more of the aortic valve masses, at the time of surgical resection, into the right coronary artery at the time of surgical resection, was also entertained as a possibility but quickly discarded given the lack of regional wall motion abnormalities in the left ventricle and the extremely unlikely chance of an isolated embolization to the right ventricular branch of the right coronary artery.
Closure of an atrial septal defect in a patient with significant pulmonary hypertension might cause some of the hemodynamic derangements seen in this patient; however, closure of a PFO with similar consequences would be mostly unlikely.
Mobilization of all pulmonary veins with application of radiofrequency lesions just proximal to their entrance may account for all the findings in our case [bib_ref] Acute fatal pulmonary vein occlusion after catheter ablation of atrial fibrillation, Nilsson [/bib_ref] [bib_ref] Pulmonary vein stenosis complicating catheter ablation of focal atrial fibrillation, Scanavacca [/bib_ref] [bib_ref] Pulmonary vein stenosis after catheter ablation of atrial fibrillation, Robbins [/bib_ref] , however, stenosis of the pulmonary veins are usually noted at a much later time and would not resolve with tricuspid valve repair.
Lastly, manual damage to the tricuspid valve during surgical manipulation has to be taken into account; however, the valve anatomy was intact and only the annulus was dilated. So in the end, pulmonary embolism (PE) remained as the most likely possibility despite the conflicting diagnostic testing results. Although helical (spiral) CT with bilateral upper extremity contrast injection has a reported sensitivity of 98 percent for diagnosing PE [bib_ref] Helical CT phlebography of the superior vena cava: diagnosis and evaluation of..., Qanadli [/bib_ref] ; in some studies the sensitivities have ranged from 53 to 87 percent, even for segmental or larger emboli [bib_ref] Acute pulmonary embolism: assessment of helical CT for diagnosis, Drucker [/bib_ref] [bib_ref] Pulmonary embolism: prospective comparison of spiral CT with ventilation-perfusion scintigraphy, Mayo [/bib_ref] [bib_ref] Performance of helical computed tomography in unselected outpatients with suspected pulmonary embolism, Perrier [/bib_ref]. Obviously, the specificity of helical CT scanning depends on reader experience and results must be interpreted with caution, particularly if the clinical probability of pulmonary emboli and the CT results are discordant. A ventilation-perfusion scan was not considered in this case given the patient's critical and unstable condition that would place the patient at an enormous risk to undergo this test. Furthermore, the results of a recent report that demonstrated that helical CT had a greater discriminatory power than ventilation-perfusion scanning with normal and/or near-normal threshold to exclude pulmonary embolism we decided not to pursue this venue, but instead consider another diagnostic modality. [bib_ref] Ventilation-perfusion scanning and helical CT in suspected pulmonary embolism: meta-analysis of diagnostic..., Hayashino [/bib_ref] It has to mentioned that the use of D-dimer in this particular case was considered of limited value given the known altered fibrinolysis and platelet function that occurs in patients post cardiopulmonary bypass resulting in an increase in soluble fibrin, fibrinogen degradation products and D-dimer. [bib_ref] Enhanced effective fibrinolysis following the neutralization of heparin in open heart surgery..., Gram [/bib_ref] [bib_ref] Relationship of fibrinolysis and platelet function to bleeding after cardiopulmonary bypass, Ray [/bib_ref] However, of particular importance and relevance to this discussion are the findings noted on echocardiography. It has been suggested that in up to more than 80 percent of patients with PE, some echocardiographic abnormality that includes either right ventricular dilatation and dysfunction or color Doppler manifestation of tricuspid regurgitation will be present in these patients using this diagnostic modality [bib_ref] Echocardiographic evaluation of pulmonary embolism and its response to therapeutic interventions, Come [/bib_ref] [bib_ref] Short-term clinical outcome of patients with acute pulmonary embolism, normal blood pressure,..., Grifoni [/bib_ref]. Thus, although we should have discarded PE on our case based on the spiral chest CT results, we continued pursuing this diagnostic entity given the unexplained dilatation of both inferior and superior Bi-caval transesophageal view showing the RA and LA divided by the interatrial septum Bi-caval transesophageal view showing the RA and LA divided by the interatrial septum. In this view the superior vena cava (SVC) is also seen. Please note, the small homogeneous mass arising from the orifice of the inferior vena cava, as seen flanked by the arrows.
vena cavae [bib_ref] Acute pulmonary embolism: value of transthoracic and transesophageal echocardiography in comparison with..., Steiner [/bib_ref]. In fact, upon careful review of the post operative TEE, a small homogeneous echodensity arising from the inferior vena cava, as seen in , was seen. This homogeneous echodensity, highly suggestive of a clot, that probably formed during closure of the PFO or less likely present on the original study; but also undetected, then migrated to the right ventricle and eventually to the main pulmonary artery at the termination of the surgical procedure and it was not responsible for the clinical events until later in the postoperative course.
Although the diagnosis of PE in this case will certainly account for the clinical findings and hemodynamic abnormalities, we still have to explain the degree of tricuspid regurgitation and the tricuspid annular dilatation noted post operatively. It has been documented, that chronic thromboembolic pulmonary hypertension is associated with right ventricular dilatation, high rightsided filling pressures, and functional tricuspid regurgitation. Although we were not able to obtain a copy of the original study, it is possible that some degree of right ventricular dysfunction was already present and was further compounded by an acute event that disrupted homeostasis and resulted in the abnormal hemodynamics. This is a likely possibility, since it has been postulated, that the tricuspid regurgitation as a result of chronic thromboembolic disease that produces pulmonary hypertension is caused by tricuspid annular dilatation with displacement of the papillary muscles [bib_ref] Tricuspid valvular disease in the patient with chronic pulmonary thromboembolic disease, Thistlethwaite [/bib_ref]. In addition, as reported by Rosenberger and associates, TEE evidence of right ventricular dysfunction and new or worsening degree of tricuspid regurgitation are found in a significant number of patients with PE even when direct visualization of a clot was not found on TEE or it was questionable. [bib_ref] Utility of intraoperative transesophageal echocardiography for diagnosis of pulmonary embolism, Rosenberger [/bib_ref] In this case, leftward interatrial septal bowing could not be used as a TEE finding since the patient was intubated and mechanically ventilated.
Finally, the decision of proceeding with tricuspid valve repair was basically made given the TEE results showing incomplete coaptation of the tricuspid valve leaflets with a clearly dysfunctional right ventricle in a patient that continue to deteriorate despite inotrope support. In a recent report by Dreyfus and associates, remodeling annuloplasty of the tricuspid valve based on tricuspid dilation has shown to improve functional status irrespective of the grade of regurgitation and these authors conclude that if left alone, tricuspid dilatation is an ongoing disease process that will, with time, lead to severe tricuspid regurgitation [bib_ref] Secondary tricuspid regurgitation or dilatation: which should be the criteria for surgical..., Dreyfus [/bib_ref] and this in turn will continue to compromise right ventricular hemodynamics.
# Conclusion
This case demonstrates an unusual post operative course of events and physical findings in which a unifying diag-nosis was difficult to identify given the conflicting nature of diagnostic testing. We present a careful retrospective examination of all clinical data and review possible clinical entities that could have been implicated in this particular case. In addition, we point out how TEE findings were most useful in identifying the best course of action. We postulate that in our case, resolution of the PE with anticoagulation [bib_ref] Massive pulmonary embolism with large floating thrombus in the truncus of the..., Maier [/bib_ref] and reestablishment of a normal right ventricular geometry with tricuspid valve repair [bib_ref] Idiopathic annular dilation: a rare cause of isolated severe tricuspid regurgitation, Girard [/bib_ref] worked in unison in restoring normal hemodynamics and resolving both dyspnea and venous dilatation.
## Abbreviations
|
Altering product placement to create a healthier layout in supermarkets: Outcomes on store sales, customer purchasing, and diet in a prospective matched controlled cluster study
BackgroundAU : Pleaseconfirmthatallheadinglevelsarerepresentedcorrectly: Previous product placement trials in supermarkets are limited in scope and outcome data collected. This study assessed the effects on store-level sales, household-level purchasing, and dietary behaviours of a healthier supermarket layout. TAU : PleasecheckwhethertheeditstothesentenceThisisaprospectivematchedcontrolledclustertrialwith:::areco his is a prospective matched controlled cluster trial with 2 intervention components: (i) new fresh fruit and vegetable sections near store entrances (replacing smaller displays at the back) and frozen vegetables repositioned to the entrance aisle, plus (ii) the removal of confectionery from checkouts and aisle ends opposite. In this pilot study, the intervention was implemented for 6 months in 3 discount supermarkets in England. Three control stores were matched on store sales and customer profiles and neighbourhood deprivation. Women customers aged 18 to 45 years, with loyalty cards, were assigned to the intervention (n = 62) or control group (n = 88) of their primary store. The trial registration number is NCT03518151. Interrupted time series analysis showed that increases in store-level sales of fruits and vegetables were greater in intervention stores than predicted at 3 (1.71 standard deviations (SDs) (95% CI 0.45, 2.96), P = 0.01) and 6 months follow-up (2.42 SDs (0.22, 4.62), P = 0.03), equivalent to approximately 6,170 and approximately 9,820 extra portions per store, per week, respectively. The proportion of purchasing fruits and vegetables per week rose among intervention participants at 3 and 6 months compared to control participants (0.2% versus −3.0%, P = 0.22; 1.7% versus −3.5%, P = 0.05, respectively). Store sales of PLOS MEDICINE PLOS Medicine | https://doi.org/10.1371/journal.pmed.Data Availability Statement: Data cannot be shared publicly because of the conditions of the agreement with the commercial collaborator. Data collected by the research team are available from the MRC Lifecourse Epidemiology Unit Data Manager ([email protected]) for researchers who meet the criteria for access to confidential data.Methods and findingsconfectionery were lower in intervention stores than predicted at 3 (−1.05 SDs (−1.98, −0.12), P = 0.03) and 6 months (−1.37 SDs (−2.95, 0.22), P = 0.09), equivalent to approximately 1,359 and approximately 1,575 fewer portions per store, per week, respectively; no differences were observed for confectionery purchasing. Changes in dietary variables were predominantly in the expected direction for health benefit. Intervention implementation was not within control of the research team, and stores could not be randomised. It is a pilot study, and, therefore, not powered to detect an effect.ConclusionsHealthier supermarket layouts can improve the nutrition profile of store sales and likely improve household purchasing and dietary quality. Placing fruits and vegetables near store entrances should be considered alongside policies to limit prominent placement of unhealthy foods. ClinicalTrials.gov NCT03518151 (pre-results) Author summary Why was this study done?Trial registration- Supermarkets are a major source of food for many families, and in-store marketing techniques like prominent product placement can influence customers' food choices.- Studies testing whether placement strategies can promote healthier food purchasing have been limited in scope, placing healthy products alongside unhealthy products or a single location (i.e., checkouts), and findings have been inconsistent.- Governments are proposing to ban the use of prominent placement strategies for unhealthy food and drink products in supermarkets, yet evidence from robustly designed real-world studies remains limited.What did the researchers do and find?- We assessed whether creating a healthier layout in discount supermarkets in England improves the healthiness of foods bought by all customers who visit the stores, as well as foods purchased and eaten by women customers aged 18 to 45 years.- We found that storewide confectionery sales declined, and fruit and vegetable sales increased, when nonfood items and water were placed at checkouts and aisle ends opposite, and an expanded fruit and vegetable section was repositioned near the store entrance; women's fruit and vegetable purchasing and dietary quality improved, but their confectionery purchasing did not.PLOS MEDICINEOutcomes of a product placement supermarket trial PLOS Medicine | https://doi.What do these findings mean?- Our results should be viewed with some caution; because of some limitations in the design of the study, the results may be over-or underestimated. However, bans on prominent placement of unhealthy foods in supermarkets could be beneficial for population diet, and effects may be even better if supermarkets were also required to place produce near their entrances.PLOS MEDICINEOutcomes of a product placement supermarket trial PLOS Medicine | https://doi.
# Introduction
Obesity and poor diet constitute 2 of the greatest threats to the population health . They are costly to societyand disproportionately affect those who are socioeconomically disadvantaged. The importance of improving population diet is ever more apparent with obesity and poor diet emerging as key risk factors for a severe response to the Coronavirus Disease 2019 (CAU : PleasenotethatCOVID À 19hasbeendefinedasCoronavirusDisease2019inthesentenceTheim OVID-19) infection, particularly among adults aged under 65 years. Most families rely on supermarkets for their food; during COVID-19 lockdown, this reliance increased. Such dependence on supermarkets as a primary food source makes them an appropriate setting for interventions to improve dietary behaviours. Despite online grocery sales increasing from 7% to 10% of the United Kingdom market recently, the majority of sales occur in-store where customers are exposed to marketing techniques that attempt to influence their food choices and preferences. Product placement is one marketing technique used in supermarkets that predominantly promotes unhealthy food and beverage choices. For example, in the UK, two-thirds of all food and drink products placed in prominent in-store locations, such as checkouts, store entrances, end of aisles, and freestanding display units, have been found to be sugary or calorie-dense, ultraprocessed products; less than 1% of food products positioned in these prominent supermarket locations were fruits or vegetables. This finding is a concern for population health, with growing evidence that these placement strategies can prompt customers to buy these unhealthy products.
In an effort to curb the influence of unhealthy marketing tactics on population diet, the UK government announced their intention to ban the use of prominent placement strategies for unhealthy food and drink products in supermarkets and other outlets. This proposal forms part of the national strategy to address childhood obesity. There is a pressing need for further evidence from local, well-designed intervention studies aimed at testing the effect, and cost impact, of healthier product placement strategies. Such evidence could assist UK policy makers appropriately frame the proposed ban, as well as help guide future government intervention to improve diet across the world.
Previous studies testing the effect of "healthier checkouts" in supermarkets, which placed healthier snack items alongside or at alternate checkouts to unhealthy snacks, have shown limited success at reducing sales of unhealthy food and beverages. Few supermarket trials have tested the effects of removing unhealthy products from all checkouts, and none, to our knowledge, have additionally tested the effects of removing unhealthy products from all the end-of-aisle displays opposite checkouts. Furthermore, little prior research has considered the impact of replacing unhealthy food products at checkout areas with nonfood items in an effort to restrict opportunities for impulsive calorie purchases while aiming to preserve impulse expenditure. Positioning products near the store entrance is another prominent placement strategy used by supermarkets to tempt customers. While many supermarkets do place fresh fruits and vegetables in a position that customers encounter when first entering the store, a number of discount and small supermarket chains do not routinely place fruits and vegetables near the store entrance. UK research shows that discount and small supermarkets have less healthy environments than other UK supermarkets, including lower availability and less prominent placement of fresh fruits and vegetables. These poorer in-store environments may be contributing to dietary inequalities because families experiencing disadvantage and younger adults, known to have poorer quality diets, frequently rely on these stores for their food. Retail intervention studies in the United States have shown promising effects on food purchasing habits among low-income, minority groups when fresh fruit and vegetable displays were moved to the front of the store. Evaluation of the effects of such a strategy, in combination with the removal of confectionery from checkouts, is needed and could inform future government policy.
Healthier product placement interventions could have economic implications for supermarkets and individuals with potential impact on commercial viability and household food shopping budgets, respectively. Economic evaluations of health-related supermarket interventions have been rarely considered, yet such information is needed to inform future policy action.
This study will help to address current evidence gaps regarding the use of prominent placement strategies to support improvements in population diet. It aims to assess whether creating a healthier layout in discount supermarkets in England improves the healthiness of store sales (primary outcome) and the purchasing and dietary behaviours of women customers aged 18 to 45 years (secondary outcomes) after 3 and 6 months. To our knowledge, this study is unique in its analysis of individual loyalty card data, in addition to store sales data, as well as collecting dietary data from more than 1 family member. The study also evaluated possible cost implications of the intervention from individual and retailer perspectives.
# Methodology
## Study design and setting
This was a pilot study with a prospective matched controlled cluster design, with participants clustered within 6 study supermarkets to account for the store-based intervention. The flow diagram, Fig A in S1 File, illustrates the time frame of store sales, participant purchasing dietary data collection. The study, which took place between April 2016 and March 2017, was approved by the University of Southampton, Faculty of Medicine Ethics Committee (ID 20986.A2) and was conducted in accordance with the Declaration of Helsinki and data protection regulations; it was registered with ClinicalTrials.gov (NCT03518151, pre-results).
The setting for this study was stores of a discount supermarket chain located in more socioeconomically deprived neighbourhoods (within the most deprived 5 Index of Multiple Deprivation (IMD) decilesacross England. The collaborating supermarket has over 900 stores nationwide and holds approximately 2% of the grocery market share in the UK.
This pilot study sampled 6 stores, 3 intervention and 3 control stores. The number of stores included was determined by the refurbishment schedule of the supermarket chain. Recruitment of additional stores and randomisation of stores were not viable within the company's business model; intervention stores were selected because structural changes to their in-store environment had already been planned at the time of the study by the company, had an average sales profile, and were located in areas with higher neighbourhood deprivation. Control stores were matched to an intervention store based on (i) sales profile; (ii) customer profile; and (iii) neighbourhood deprivation (IMD). Matching on these factors aimed to increase the similarity of intervention and control stores and reduce the effects of confounding. Control stores were geographically distant from intervention stores to reduce contamination effects of control women shopping at intervention stores.
## Intervention and control conditions
The intervention was implemented continuously for 6 months and had 2 components executed simultaneously: (i) more prominent placement of fruits and vegetables; and (ii) removal of unhealthy foods from checkouts and the end-of-aisle opposite checkouts. The first intervention component involved expanding the produce section to increase the availability of fresh fruits and vegetables and positioning the produce near the store entrance. Frozen vegetables were also relocated to the first aisle, a more prominent position in store. All unhealthy foods (confectionery, crisps, biscuits, etc.), but predominantly confectionery (chocolate, sweets, or candy), were removed from all checkouts and displays at the end-of-aisle opposite checkouts and replaced with nonfood items (i.e., tissues, painkillers, lip balm, cleansing wipes, toothpaste, soap, deodorant, and hand wash), water, and sugar-free gum. One intervention store also positioned some fresh fruits and vegetables at the checkouts because of the size and shape of the checkout display unit. In each intervention store, the confectionery section was moved to the least prominent position, the last aisle of the store. Seasonal confectionery (i.e., Easter, Christmas, Mother's/Father's Day, Valentine's Day, and Halloween) and branded confectionery for which marketing space was already paid was positioned at the store entrance and in freestanding displays in the aisles but not at the checkouts or end-of-aisle opposite checkouts during the intervention period. Intervention stores also underwent improvements in presentation (e.g., cleaning, painting, and updated signage) at the time the intervention was implemented.
The control condition was the previous layout of stores, as at the baseline period, with (i) a limited range of fresh fruits and vegetables, placed at the back of the store; (ii) frozen vegetables placed in a middle aisle of the store; and (iii) confectionery placed at the checkouts and the confectionery section positioned at end-of-aisle opposite checkouts.
## Participant eligibility and recruitment
Women of childbearing age were targeted for this intervention because of their role as household food gatekeepersand their influence on the short-and long-term health of the next generation. For the children of these women, establishing good dietary habits early in life is important for optimal growth, development, and long-term health. Eligible participants were women, aged 18 to 45 years, who held a loyalty card with the study supermarket chain and had shopped in a study store in the 12 weeks before recruitment (according to loyalty card data). Women under the age of 18 or over 45 years at the time of the study who did not hold a loyalty card or only shopped online were not eligible to participate. Recruitment occurred in 3 waves between July and September 2016. Women from each pair of stores were recruited over the same period, prior to the implementation period for that intervention store. Eligible women in all 6 study stores, identified from the loyalty card register, were sent a letter inviting them to participate in a study that was investigating the food shopping and eating patterns of women aged 18 to 45 years. The letter did not contain details about the intervention. The letter was sent by the supermarket on behalf of the research team in order to comply with data protection laws. Interested women contacted the research team directly via freephone number, text, or email and were screened for eligibility and then provided informed consent.
In addition to being mailed a letter, participants in the first pair of stores were initially contacted by the supermarket via email and text message, and advertisements about the study were placed on the back of shopping receipts and on Facebook. These additional recruitment methods, however, yielded very little interest from participants and were thus phased out over the duration of the study. In order to boost participant numbers in the first pair of stores, in-store recruitment was used, whereby members of the research team approached women customers while shopping and provided them with a study information sheet. Interested women registered with the researcher in-store and were subsequently phoned and consented. This method proved effective at enhancing representation of disadvantaged customers and was used for all 6 study stores. To promote retention, all participants were offered up to 3× £10 Love2Shop vouchers for taking part in the study. For comparison, the national minimum living wage rate for adults over 25 years in 2016 was £7.20/hour. Love2Shop vouchers are multioption vouchers that can be used at 150 leading high street retailers, which span a range of retail categories.
## Outcome measures
The data collected included 9 months continuous store sale transaction data (primary outcome) and participant loyalty card data (secondary outcome) provided by the collaborating supermarket to cover 3 time periods: time (1) baseline (3 months prior to intervention implementation; time (2) short-term intervention effects (0 to 3 months postintervention commencement because evidence suggests that this represents an appropriate interval for habit formation; and time (3) longer-term intervention effects (3 to 6 months postintervention commencement to assess sustained behavioural changes). In order to obtain an understanding of intervention effects on household members' diets (secondary outcomes), interview-administered telephone questionnaires obtained information about participants' diets and diets of their children aged 2 to 6 years (where applicable) at 3 time points: baseline (time 1) and 3 (time 2) and 6 months (time 3) following intervention commencement.
Store sales of fresh fruits and vegetables, frozen vegetables, confectionery, and intervention checkout items were provided as numbers of items for each product sold in each week of the study period. Participant purchasing data covering the same categories were provided as the number of items for each product purchased at each store visit during the study period. The research team aggregated these purchasing data from each visit to a weekly structure for analysis to enable our data to be presented as items per household per week in order to be able to detect changes in visit frequency as well as purchasing quantity, a method used in previous supermarket trials. Store closure for structural changes to intervention stores affected 2 weeks of sales and purchasing data; these 2 weeks of data were removed from the analysis for both the intervention and matched control stores. Christmas notably impacted another 2 weeks, requiring a further 2 weeks of data to be removed. Store sales and individual purchasing datasets consisted of the 11 weeks prior to the intervention and 24 weeks afterward.
Measures of women's dietary quality, and their young children's dietary quality (where relevant), were assessed using published, validated tools. Participants were asked to indicate how often in the previous month they (or their child) consumed each of 20 foods in a Food Frequency Questionnaire (FFQ). A dietary quality score for each woman (or child) was calculated by multiplying their reported frequency of consumption of each of the 20 items from their FFQ by corresponding weightings derived from the published tools (based on principal component analysis) and then summing the results. Dietary quality scores were then standardised to have a mean of 0 and standard deviation (SD) of 1. Higher scores represent better dietary quality characterised by higher intakes of vegetables, fruit, water, and whole grain bread and lower intakes of white bread, processed meats, chips, crisps, and sugar. Women's daily fruit and vegetable intake was measured using a 2-item tool. This measure details the amount (quantity) of fruits and vegetables eaten and complemented the frequency data collected by the FFQ. TAU : ThesentenceThefinancialeffectsoftheintervention:::isincomplete:Pleaseupdateandcorrect:
he financial effects of the intervention on stores and women was assessed by calculating changes in total weekly store sales and changes in the amounts of money participants spent on grocery foods per week respectively, from before to after the intervention. Participants reported, at each survey wave, the total amount of money they spent on groceries in the past month. Participants' weekly household spend on groceries at the collaborating supermarket chain were provided by the loyalty card data, and weekly total sales for each store were provided by the store transaction data.
## Fidelity assessment
All stores were visited by a member of the research team during the baseline period prior to intervention implementation to assess whether the preintervention and control layouts were similar for each pair of stores. Post-intervention visits and phone calls were made to all stores to assess fidelity of both control and intervention conditions using photographic monitoring and discussions with supermarket staff.
# Statistical analysis
Descriptive variables are given as percentage (frequency) for categorical variables and median (interquartile range) for nonnormally distributed continuous variables. Differences between intervention and control participants were tested using chi-squared tests for categorical variables and Mann-Whitney rank sum tests for nonnormally distributed continuous variables.
The distributions of the data were unknown in advance of study commencement because this was a pilot study. The research team therefore made decisions about the statistical analyses following (i) completion of initial analyses to determine the shape of the outcome data; and (ii) consultation with the literature to identify appropriate statistical methods for the various outcome variables.
Store sales data were analysed using an interrupted time series. Weekly sales per store of fresh fruits and vegetables, frozen vegetables, confectionery, and intervention checkout items were transformed to normal distributions using inverse normal (also known as Fisher-Yates) transformationsto protect commercially sensitive sales figures. Time series models were fitted with terms for study week (linear term and weeks from baseline), intervention, level (an indicator of the postintervention period), trend (study week in the postintervention period), and interactions between intervention and study week, intervention and level, and intervention and trend. By including the variable "level," the model will test for a step change at the time of the intervention because this immediate effect was anticipated based on conceptual models of consumer behaviour and evidence from existing supermarket placement studies. The time series models were fitted separately in each pair of stores in order to account for the store pairing in the analyses. For the confectionery outcome, there was a strong pattern of steady increase in sales until Christmas, an abrupt fall in sales over the Christmas period, and a steady increase in sales after Christmas; therefore, an additional term for post-Christmas level was included in the confectionery models. The P value for the interaction between intervention and level indicates the significance of the impact of the intervention on level of store sales at the time of the intervention. A counterfactual line is included on the interrupted time series graphs, indicating the trends in sales that would have been expected had the intervention not occurred. Confidence intervals at 3 and 6 months postintervention were calculated using the delta method. In order to inform planning of future cluster trials, we also fitted a random effects multilevel linear regression model in order to calculate an intraclass correlation coefficient (ICC). The outcome variable was baseline weekly sales per store of fresh fruit and vegetable z-score, and store ID was used to define clustering.
Fixed-effects meta-analysiswas used to synthesise the differences between pairs of stores at the time of intervention, 3 and 6 months postintervention. This method was selected to enable the (i) retention of the store pairing in the study design; (ii) comparisons between pairs; and (iii) overall statements of study effect size and precision. Results were interpreted on the original items sold per week scale by calculating the equivalent change on the original scale to the change from the median on the Fisher-Yates transformed scale. The collaborating supermarket chain sells only packaged fruits and vegetables (products were not sold singly), with each item averaging 5 portions (approximately 400 g). Similarly, confectionery sales data indicated that the most popular items weighed 100 to 200 g. Thus, applying national recommendationseach item was considered equivalent to approximately 3 portions (150g); this portion size information was used to convert results from items to portions.
For the individual purchasing data, it was not possible to use a time series analysis because the data had a strong right-hand skew; for example, 83% of the women's weekly purchases resulted in no sales of fresh fruits and vegetables. The outcome data were therefore dichotomised to indicate whether each week resulted in any purchases of the food category under consideration. A difference-in-difference approach was used, in line with previous supermarket placement research, so that each logistic regression model included fixed effects for intervention group, time period, and the interaction between intervention group and time period. Time period was coded as 2 dummy variables indicating the 0 to 3 and 3 to 6 months periods postintervention. The interaction terms test whether the difference between purchasing during the intervention compared to during the preintervention period differed between intervention and control stores. In addition, random effects were included for women, to account for the multilevel structure of the data, with weeks clustered within women; there was an insufficient number of stores in this study to include random effects for stores. Women's data were analysed according to the store they were recruited from in order to conform to an intention-totreat analysis.
The effects of the intervention on changes in diet from baseline to 3 and 6 months postintervention were explored using linear regression models with diet as the outcome and intervention group, diet at baseline, and IMD (to control for similarities between pairs of stores) as predictors. A second set of regression models included confounding variables determined prior to analysis using a directed acyclic graph (DAG) (Figs B and C in S1 File); variables included as confounding variables were age, money spent on groceries, number of children in the household, and woman's education.
Changes in amounts of money spent on grocery foods per week were analysed using a difference-in-difference approach. The distribution of the amounts of money spent on grocery foods was right-skewed, so a log transformation was applied. A linear regression model was fitted including effects for intervention group, time period, and the interaction between intervention group and time period. Time period was coded as 2 dummy variables indicating the 0-3 and 3-6 month periods postintervention. The interaction terms test whether the difference between the amounts of money spent on grocery foods during the intervention compared to during the preintervention period differed between intervention and control stores. Total individual purchasing data were similarly analysed using a difference-in-difference approach based on a linear regression model with average weekly purchases in pounds (£) as the outcome and interaction terms (as described for the reported money spent on grocery foods). Total stores sales data were analysed using an interrupted time series in the same way as the confectionery data (i.e., an additional term for post-Christmas level was included in the models).
All analyses were performed in Stata 14, except the time series models that were fitted in R. In line with current statistical thinkingand due to pilot nature of this study, we base our interpretations on effect sizes, and their precision, rather than emphasising the statistical significance relative to any boundary. We also provide P values for transparency.
# Results
## Participant characteristics
A total of 150 women aged 18 to 45 years who regularly shopped at 1 of the 6 study stores were recruited; 121 participants reported living with children (aged <18 years), and 72 of these participants also provided data about their child aged 2 to 6 years. The most successful recruitment methods were posted letters (49%) and in-store recruitment (34%). Attrition rates were low: 8% at 3-month and 14% at 6-month follow-ups. There were no significant differences in participant characteristics at baseline between intervention and control participants. Participants' median age was 36 years, 91% were white British, 59% had low educational attainment (no qualifications beyond age 16 years), and 46% were in paid employment. Almost half reported that the collaborating supermarket chain was where they purchased most of their groceries (44%).
## Store sales
The average number of items sold per week per store (in SDs) in store pairs is shown in Figs D and E in S1 File (complete data from all 6 stores); the synthesised results using meta-analysis are shown in
## Individual purchasing
Of the 150 participants, 107 women (1,539 visits) had individual purchasing data from their loyalty cards for the study period, 56 recruited from control stores (851 visits) and 51 from intervention stores (688 visits). These missing data relate to the need for explicit consent from participants to the supermarket for them to legally share their loyalty card data; not all participants completed this additional consenting step. Of the 1,539 visits, 253 (140 control women visits and 113 intervention women visits) were not at the stores the women were recruited from. All 253 visits to alternative stores were to nonstudy stores, meaning that at 113 visits, the intervention women were not exposed to the store changes. Modelled proportions of women purchasing food items are shown inThe proportion of purchasing fresh fruits and vegetables per week rose in intervention stores 3 months postintervention (0.2% (95% CI −3.6%, 3.9%)) compared to a drop in control stores (−3.0% (−6.6%, 0.6%)) (P = 0.22); even greater changes were seen 6 months postintervention (1.7% (−2.2%, 5.6%) in intervention stores compared to −3.5% (−7.2%, 0.1%) in control stores, P = 0.05). Frozen vegetable purchasing patterns demonstrated a decrease from baseline to 3 months postintervention among women in both intervention (−2.1% (95% CI −4.9%, 0.6%)) and control groups (−0.1% (95% CI −3.9%, 0.2%)) (P = 0.34) and from baseline to 6 months postintervention among women in both intervention (−3.0% (95% CI −6.1%, 0.1%)) and control groups (−3.1% (95% CI −5.9%, 0.3%)) (P = 0.53). Purchasing patterns for confectionery was also similar among women in intervention and control groups showing an increase from baseline to 3 months (P = 0.94) and 6 months (P = 0.71) postintervention. The proportion of women purchasing intervention checkout items were much lower in the intervention than control stores at baseline, but rose among intervention women by 1.0% (95% CI −0.9%, 2.8%) and 0.6% (−1.3%, 2.4%) in the 3 and 6 months postintervention periods, respectively, compared to a fall of −1.8% (−3.8%, 0.2%) and −1.6% (−3.6%, 0.5%) among control women (P = 0.04 and P = 0.13), respectively.
## Dietary variables
Differences in women's and children's diets over time between groups are presented in. After adjustment for IMD, diet quality score at 3 months follow-up increased from baseline by 0.29 SDs ((95% CI 0.01, 0.57), P = 0.04) among intervention women compared to those shopping in control stores. This difference is equivalent to approximately 6 additional Women's reported daily intake of fruits and vegetables increased, among intervention compared to control women, from baseline to 3 and 6 months follow-up, with greater increase observed at the later follow-up. Reported weekly confectionery intake decreased from baseline to 3 months follow-up among intervention women, compared to control women, but increased from baseline to 6 months follow-up.
## Economic impacts
There was a relatively small difference in the mean (SD) change from baseline to 3-
## Fidelity assessment
The fruit and vegetable components of the intervention were fully implemented in all intervention stores. Stock levels of nonfood items at checkouts were reported and observed to be lower than anticipated in 2 of the intervention stores during the 3-6-month postintervention period. These issues were attributed, at least in part, to demand outweighing supply of the designated intervention checkout items.
# Discussion
## Principal findings
This pilot supermarket trial showed that creating a healthier store layout by expanding the range of fruits and vegetables and placing them near the entrance, plus removing all unhealthy foods, namely confectionery, from checkouts and aisle ends opposite checkouts had a positive effect for health benefit, increasing fresh fruit and vegetable sales and reducing confectionery sales at a population (store) level. Among a sample of women customers of childbearing age, the intervention showed beneficial effects on fresh fruit and vegetable purchasing patterns, particularly after 6 months of intervention implementation, but no intervention effect was observed for participants' confectionery purchases. Nonfood items, water, and sugar-free gum, which were placed at checkouts during the intervention, were purchased more by trial participants exposed to the intervention, but did not translate into increased sales of these items at the store level. Findings from assessment of the impact of the intervention on women's and children's diets were more equivocal. Trends for the dietary variables were predominantly in the expected direction for health benefit, with improvements in women's overall dietary quality and daily fruit and vegetable intake at 3 months follow-up, particularly noteworthy for size of dietary change. The economic analysis showed virtually no impact on weekly household grocery spend across all participants or overall weekly store sales across all stores, indicating no detrimental cost effect of the intervention to participants or the retailer.
# Strengths and limitations
To our knowledge, this is the first health-focused product placement study in supermarkets to use (i) loyalty card data to track effects on household purchasing patterns of existing customers in addition to store sales data to measure population-level effects; and (ii) dietary data from more than 1 household member to aid understanding of who has been affected by the intervention. This study has a number of other advantages over existing supermarket placement researchincluding use of a matched comparison group; use of robust statistical analysis methods, particularly the interrupted time series approach to assess intervention effects on store sales; and the study setting and sample including a high representation of families from lower socioeconomic backgrounds, thus providing important information on interventions that have potential to reduce inequalities. The intervention included both reduced availability and prominence of unhealthy foods and improved availability and prominence of fruits and vegetables, plus it trialled the use of nonfood items at checkouts in an effort to maintain business profitability while reducing customers' calorie purchasing opportunities.
This study has a number of limitations. It was not possible to randomise stores to intervention or control groups; thus, the study may be biased by unmeasured confounding effects. Parallel designs, like the one used in our study, with control groups matched on area characteristics and store sales (plus adjustment for confounders), however, do offer a robust design in realworld settings and enables valuable knowledge of intervention effectiveness in complex social contexts to be shared with policy makers, particularly with data collected at store, household, and individual levels. Although the number of stores and participants that took part in this study is consistent with previous placement intervention research, this was a pilot study and is underpowered because store-based placement intervention studies require power calculations that take account of clustering at the store level, so the study results should be viewed with some caution. Nevertheless, the study's effect sizes demonstrated meaningful improvements in the healthiness of store sales and in household purchases of fresh produce.
Store selection and intervention implementation were not within the control of the research team, and some issues were identified. Under-or overestimation of intervention effects observed may therefore be possible. Finally, the economic analysis was limited in scope and did not include broader cost implications such as time or travel costs for individuals or profit loss, infrastructure, and staff training costs for the retailer, nor were the benefits of improved dietary quality on health or well-being calculated.
## Comparison with previous literature
Previous intervention studies that repositioned fresh fruits and vegetables to prominent locations in food stores have found effects in a similar direction, but smaller in magnitude than the potentially meaningful effect sizes observed in our study. One study set in discount supermarkets in Denmark failed to demonstrate significant intervention effects on store sales. This finding may relate to the intervention involving prominent positioning of additional produce bin displays rather than repositioning of the entire produce section near the store entrance and the shorter 3-month intervention duration.
Our study showed stronger intervention effects at 6 months than at 3 months follow-up, suggesting that changes in food shopping, like other health behaviours, take approximately 14 weeks for habit formation and strengthen over time. Another study, in convenience stores in the US, which moved the fresh produce section to intervention stores' entrances, found that fruit and vegetable sales from government assistance vouchers for young, low-income families increased in intervention stores during the 5-month intervention, yet decreased in control stores (difference of US$63/month between groups). Women's fresh fruit and vegetable purchasing patterns from loyalty card transactions in our study similarly revealed increasing and decreasing trends for intervention and control participants, respectively; placing the produce section in a prominent in-store position may have protective benefits as indicated by the improvements in intervention women's dietary quality. This finding could be of particular importance because long-term analyses of UK purchases of fresh fruits and vegetables show a general decline: Fresh green vegetable purchases dropped 25% between 2005 and 2015, and fresh fruit purchases decreased by 17% from 2006 to 2015 [51].
Our finding, that the intervention had a beneficial effect at reducing confectionery sales at the store level, differs in strength of effect from previous checkout intervention research from the Netherlands, US, Denmark, and Norway. This difference may be attributable to the fact that previous interventions positioned healthier products either alongside or at alternative checkouts to existing unhealthy foods or beverages, rather than removing them completely. Additionally, our study also removed confectionery from the aisle ends opposite checkouts. This extra intervention component has not been tested previously and may have helped to enhance store-level intervention effects because more than two-thirds of products positioned in this location in UK supermarkets are unhealthy foods and beverages.
Despite the healthier design of our intervention compared to others, it did not influence the confectionery purchasing and intake patterns of women participants. Intense marketing of confectionery products occurs throughout supermarkets, not just in the checkout area, and it is probable that our study's participants were tempted by the additional positioning of confectionery products at the store entrance and freestanding aisle displays. National household purchasing trends reveal that confectionery spending continues to increase, rising 10% from 2012 to 2015 and a further 6% from 2016 to 2018. Many national celebrations have become symbolised by confectionery. During the course of our study's follow-up period, our participants celebrated Halloween, Christmas, Valentine's Day, and Mother's Day. While Easter did not fall within our study period, Easter-related treats are on sale from early January and would have been available for our participants to purchase. Research commissioned by the Royal Society for Public Health found strong customer opinions against sustained marketing of seasonal confectionery, with two-thirds of people agreeing that special occasions are used too much to advertise and sell unhealthy food and one-third stating that this strategy prompts them to have poorer diets than they normally would.
The inclusion of frozen vegetables as part of the healthier store layout intervention was novel, and, although our results did not indicate a beneficial effect on store sales or customer purchasing, no detrimental substitution effects were observed. Prominent positioning of lower calorie frozen ready meals has shown to increase sales of these products, suggesting that future research could further test the effect of enhancing placement of frozen vegetables. Such studies may be particularly pertinent when food system shocks, such as no-deal Brexit or COVID-19 seasonal agricultural workforce concerns, have the potential to hinder fresh produce availability.
## Implications for policy
In response to COVID-19, international experts are calling for strong government leadership to recalibrate economic activities in the food system for public health and the public good.
A positive example of political action includes the British Prime Minister, Boris Johnson, declaring he now supports a more interventionist approach to tackle obesity in the wake of his experience of contracting COVID-19, attributing his need for intensive care to be, at least partly, due to his being overweight. Our findings can thus provide a national, contextually relevant contribution to public policy, tentatively indicating that the planned ban on the use of prominent placement strategies for unhealthy food and drink products in supermarkets and other outlets could reduce sales of unhealthy foods like confectionery at a population level. Moreover, our findings suggest that expanding the regulation to include the placement of the produce section near store entrances in all supermarkets could increase fruit and vegetable sales, including among more disadvantaged families, and further enhance the potential benefits of the regulation to population diet. The regulation may be even more effective at limiting impulsive calorie sales if only nonfood products were to be sold in certain prominent in-store locations. Our study findings suggest that nonfood products are purchased more by customers when positioned at checkouts in place of confectionery, but adequate variety and sustained supply of nonfood products needs careful consideration to ensure viability for the retailer. This represents an important question for future research, alongside conducting a full economic evaluation from the individual, retailer, and society perspective.
## Directions for future research
The body of evidence would benefit if future intervention research studies could be adequately powered; such studies are rare and require considerable commitment from retailers. Novel trial designs using routinely collected data (e.g., loyalty card transactions within the boundaries of data protection regulations) or longitudinal observational studies may be appropriate alternatives to measure policy directives or other natural experiments. Future research might also consider assessing the independent and additive effects of altering products at different prominent in-store locations and the effects among different population groups to enhance understanding of the most effective strategies and for whom. Qualitative research with category managers who make decisions about prominent product placement could help identify existing food system drivers for unhealthy foods and opportunities/plans for changing to nonfood or healthier products, particularly in the context of looming government legislation.
# Conclusions
Although a pilot study, these results provide a comprehensive assessment of an intervention to create a healthier store layout in discount supermarkets by presenting effects on store sales, customer loyalty card purchasing patterns, and the diets of more than 1 household member. The results showed a reduction in confectionery, and increase in fresh fruits and vegetables, store sales when nonfood items and water were placed at all checkouts and aisle ends opposite, and an enhanced produce section was repositioned to near the store entrance. Beneficial effects were also observed for household fruit and vegetable purchasing and women's dietary quality. This study therefore provides novel evidence to suggest that the intended UK government ban on prominent placement of unhealthy foods across retail outlets could be beneficial for population diet and that effects may be further enhanced if requirements for a produce section near supermarket entrances were incorporated into the regulation.
Supporting information S1 File. Supporting information figures. DAG for the causal relationship between intervention and women's dietary change. DAG for the causal relationship between intervention and child's dietary change. Interrupted time series for store pairs of total sales of fresh fruits and vegetables and frozen vegetables. Interrupted time series for store pairs of total sales of confectionery and intervention checkout items by store status. DAG, directed acyclic graph. (DOCX) |
Enterococcus faecalis Endocarditis and Outpatient Treatment: A Systematic Review of Current Alternatives
The selection of the best alternative for Enterococcus faecalis infective endocarditis (IE) continuation treatment in the outpatient setting is still challenging. Three databases were searched, reporting antibiotic therapies against E. faecalis IE in or suitable for the outpatient setting. Articles the results of which were identified by species and treatment regimen were included. The quality of the studies was assessed accordingly with the study design. Data were extracted and synthesized narratively. In total, 18 studies were included. The treatment regimens reported were classified regarding the main antibiotic used as regimen, based on Aminoglycosides, dual β-lactam, teicoplanin, daptomycin or dalbavancin or oral therapy. The regimens based on aminoglycosides and dual β-lactam combinations are the treatment alternatives which gather more evidence regarding their efficacy. Dual β-lactam is the preferred option for high level aminoglycoside resistance strains, and for to its reduced nephrotoxicity, while its adaptation to the outpatient setting has been poorly documented. Less evidence supports the remaining alternatives, but many of them have been successfully adapted to outpatient care. Teicoplanin and dalbavancin as well as oral therapy seem promising. Our work provides an extensive examination of the potential alternatives to E. faecalis IE useful for outpatient care. However, the insufficient evidence hampers the attempt to give a general recommendation.. Study selection flowchart.
## Overview of the studies
As shown in, the study designs of the 18 selected articles included randomized clinical trials (n = 1), non-randomized clinical trials (n = 1), prospective cohort studies (n = 2), retrospective cohort studies (n = 9)and case series studies (n = 5). Only 10 of them analyzed more than one treatment option. The study clinical settings were outpatient (n = 4), inpatient (n = 9)and mixed (n = 5), and the therapy indication included continuation therapy in nine studies. The majority of studies included left-and right-sided endocarditis, with the exception of three studies treating only left-sided IE. There were large variations regarding the follow-up period after the ending of the antimicrobial therapy, ranging between no follow-upand one year after. Mortality was the most common outcome measure, being described in all articles, while relapse incidences were reported in 13 studiesand treatment failure in only 11 studies. Two articles presented data from the same study population; nevertheless, both were included because different aims and outcomes were assessed.
## Quality of the studies
The methodological quality of the studies included in this review was variable. Five studies were evaluated as good quality or low risk of bias, 11 studies were assessed as having some concerns or moderate risk of bias and 2 studies were reported as poor quality or serious risk of bias. The detailed results for the risk of bias assessment are summarized in.
## Therapeutic alternatives
The therapeutic alternatives identified in the articles were grouped into six categories corresponding with the main antibiotic used. The study details are summarized in.
## Overview of the studies
As shown in, the study designs of the 18 selected articles included randomized clinical trials (n = 1), non-randomized clinical trials (n = 1), prospective cohort studies (n = 2), retrospective cohort studies (n = 9)and case series studies (n = 5). Only 10 of them analyzed more than one treatment option. The study clinical settings were outpatient (n = 4), inpatient (n = 9)and mixed (n = 5), and the therapy indication included continuation therapy in nine studies. The majority of studies included left-and right-sided endocarditis, with the exception of three studies treating only left-sided IE. There were large variations regarding the follow-up period after the ending of the antimicrobial therapy, ranging between no follow-upand one year after. Mortality was the most common outcome measure, being described in all articles, while relapse incidences were reported in 13 studiesand treatment failure in only 11 studies. Two articles presented data from the same study population; nevertheless, both were included because different aims and outcomes were assessed.
## Quality of the studies
The methodological quality of the studies included in this review was variable. Five studies were evaluated as good quality or low risk of bias, 11 studies were assessed as having some concerns or moderate risk of bias and 2 studies were reported as poor quality or serious risk of bias. The detailed results for the risk of bias assessment are summarized in.
## Therapeutic alternatives
The therapeutic alternatives identified in the articles were grouped into six categories corresponding with the main antibiotic used. The study details are summarized in. The clinical outcomes of E. faecalis IE patients treated via an initial therapy with a gentamycin-based regimen were evaluated in six studies, comprising 343 episodes, none of which were in OPAT. The main combination employed was ampicillin 2 g each 4 h plus gentamicin 3 mg/kg/day during 4-6 weeks of treatment, although vancomycin and penicillin G were alternatives to ampicillin in two cases. Among the studies, adverse events rates were high, even superior to 40%, mostly due to renal toxicity. On the other hand, relapse and mortality rates oscillated between 3% and 11% and 17% and 35% respectively, after 6-12 months of minimal follow-up. One study limited patient inclusion based on IE type. This study compared the efficacy and safety of short (2 weeks) and long (4-6 weeks) treatment with gentamycin 3 mg/kg/day combined in both cases with 4-6 weeks of ampicillin or penicillin. Renal toxicity observed was significantly lower when the gentamycin short-course was used irrespective of clinical efficacy, which remained similar in both groups (69% vs. 66% of 1 year event-free survival). Likewise, Fernandez-Hidalgo et al.analyzed a large cohort of patients treated with ampicillin plus gentamycin for 4-6 weeks, in which the overall mortality after a median follow-up of 11 months was 25% and the side effects reported were 44%.
## Dual β-lactam regimens
Nine studies gathered clinical results about 337 E. faecalis IE episodes treated with dual β-lactam therapy, three of which were conducted in OPAT, and then collected data from 11 patients. The follow-up comprised six months (n = 6), three months (n = 1), and another study referred no follow-up after treatment ending. The most common combination assessed was ampicillin 2 g each 4 h plus ceftriaxone 2 g each 12 h for 4-6 weeks as an initial therapy. However, one studyadjusted it to their OPAT program, administering ceftriaxone 4 g in single daily dose, and another oneadministered ampicillin at a dose of 2 g each 6 h. Two studies proffered penicillin G as an alternative to ampicillin in a small cohort (n = 7) without any deaths or relapses. Embracing all data reported, the mortality rate was between 20% and 30% in four studies, and lower than 20% in another four. Only one study reported a mortality rate higher than 30%, maybe related to the lower dose of ampicillin administered. Regarding toxicity, the highest adverse events rate was reported by Pericas et al., showing 34% of renal failure in the cohort, although only 3% of discontinuation due to toxicity. Suzuki et al.described adverse effects in one of the four patients reported (25%), while the percentage of adverse events reported was lower than 20% in the other seven studies. In the largest cohort evaluated in the field(n = 159), the overall mortality rate was 26%, along with a low adverse events rate (9%), despite poorer general conditions than in the comparator group. Across the studies, the main conclusion was that dual β-lactam therapy with ampicillin is a primary option for E. faecalis IE, while penicillin G could be an alternative.
## Teicoplanin-based regimens
A total of 56 patients treated with teicoplanin were reported, most of them (89%) as continuation or salvage therapy. Two studies compromised episodes treated in OPAT. The largest study conducted embraced the use of monotherapy with teicoplanin for treating E. faecalis IE as continuation therapy. The reported mortality related to IE was low (8%), but the population treated with teicoplanin suffered from less severe IE than the standard therapy group.
Overall, the dose regimens were highly variable among the studies. Two of them introduced a loading dose, but the mean maintenance dose varied between 5.8 and 10 mg/kg/day, while in the other one, fixed doses were used. Within the patients treated with teicoplanin as a continuation or salvage therapy, 16 died (32%) in a minimal follow-up period of 3 months. Only three relapses were reported in these studies. The comprehensive conclusion of the studies was that teicoplanin could be an alternative for the sequential treatment of this syndrome.. The treatment scheme was considerably heterogeneous, included initial and salvage therapy, monotherapy and combine regimens, and the mean doses ranged between 8.5 and 10.125 mg/kg/day. Mortality rates reported were low (0-22%), although only one studyattained more than a one-month follow-up. Ceron et al.included OPAT treatment and described the salvage treatment of 5 E. faecalis IE episodes, of which four needed a treatment change due to treatment failure. So, the stated final conclusions differed, with two supporting daptomycin as an alternative treatment in this scenario, and one showing some concerns.
[formula] [19] SQAT Y Y Y Y Y Y Y NA Y GOOD [35] Y Y NR Y Y Y Y NR Y GOOD [24] Y Y Y Y Y Y Y NA Y GOOD [25] Y Y Y Y Y N Y NA N [/formula]
## Dalbavancin regimens
Two articlesdisclosed the outcomes of seven E. faecalis IE patients treated with dalbavancin, six (86%) with OPAT. The dosage regimen and length of the therapy encompassed single and multiple variable doses for 1 to more than 6 weeks. In these cohorts, six patients were successfully treated with dalbavancin in OPAT, hence it was proposed as an alternative.
## Oral therapy
Oral therapy has been evaluated in one large randomized clinical trial. This study comprised 400 left-side IE episodes, but it should be noted that only 20% of patients screened were included. Among them, continuation treatment with outpatient oral therapy and inpatient intravenous therapy were settled on in 201 and 199 patients, respectively. In total, 51 E. faecalis left-side IE episodes were enclosed in the oral arm. The mean time from diagnosis until the beginning of continuation of oral therapy was 17 days. High variability was shown among the antibiotic selection against E. faecalis, with six different regimen options used. Amoxicillin was part of the treatment in 90% (46/51) of these oral regimens given. The primary outcome was a composite endpoint that encompassed all-cause mortality, unplanned cardiac surgery, clinically evident embolic events or relapse of bacteremia. Considering patients with IE caused by E. faecalis treated with oral therapy, it occurred in four (7.8%), comprising only one death (1.9%).
# Discussion
This review included treatment alternatives for E. faecalis IE suitable for the outpatient setting, including oral or OPAT treatment. To this end, little evidence was found to support optimal continuation treatment. In fact, outpatient treatment of these patients has been a matter of discussion, and there is no consensus regarding patient and treatment selection. To the best of our knowledge, there is no previous work summarizing the current evidence related to E. faecalis IE treatment appropriate for outpatient care. The two main findings were the lack of randomized clinical trials and large observational studies, and the heterogeneity regarding methodology and follow-up period among the included studies. Despite this, the alternatives analyzed in this review provided a wide view on the outpatient continuation treatment for this infection.
On the one hand, the use of aminoglycosides to synergize with cell wall-active agents in the initial treatment of E. faecalis IE has been the first choice treatment for decades, in defiance of the high rates of nephrotoxicity observed. Lately, doubts about its clinical utility have been raised. This review included six studies in which this treatment option has been used as standard therapy. That said, the toxicity of prolonged treatment, and the subsequent need for drug monitoring, have been obstacles for OPAT incorporation. In this sense, a combination regimen based on a reduced gentamicin course interesting for OPAT purposes was previously proposedand thereafter deeply studied. They found a greater decrease in nephrotoxicity compared to the long gentamicin course, without repercussions in efficacy rates. The reduction in toxicity and regimen complexity are advantageous properties for OPAT inclusion. However, none of these studies have been carried out in OPAT, and also these results are only suitable for non-high-level aminoglycoside resistance (non-HLAR) strains, so its applicability could be reduced.
On the other hand, dual β-lactam therapy has been raised as a noteworthy alternative for initial treatment in the field. The safety and efficacy of this combination was substantiated in a non-randomized clinical trialand a large prospective observational study, followed by several retrospective cohort studies that confirm their findings. Despite there being enough evidence to support the use of the dual β-lactam combination in inpatient treatments, its adaptation to OPAT is complex and has been poorly studied.
Several options should be considered. The first one could be to continue using the intra-hospital regimen of ampicillin 2 g each 4 h administered through an electronic pump plus ceftriaxone 2 g each 12 h. In order to avoid the unfeasible use of two pumps simultaneously, this regimen would require the continuous assistance of a caregiver, or patient autonomy for venous access manipulation. That option is not suitable for all types of OPAT organizations, and it would remarkably reduce the patient candidates for it. Similar difficulties are found in the alternative dual β-lactam regimen, where ampicillin has been replaced by penicillin G due to discrepancies in ampicillin solution stability. Moreover, the small population studied and the lack of synergistic activity studies for the penicillin G plus ceftriaxone combination lessen its utility. For a long time, ampicillin stability data were contradictory, whereas recent studies have solved the disagreement. Besides, ampicillin and ceftriaxone daily doses diluted in the same solution were found to be stable, which could be considered as an alternative to administration through a single pump. Another suggested alternative is to switch the ceftriaxone dose regimen to 4 g in single daily doses. Four patients were treated with this combination with successful outcomes. Nonetheless, together with the small number of patients included, a serious concern about this modification is that optimal ceftriaxone exposure to achieve synergistic concentrations with ampicillin has not been described. Altogether, dual β-lactam adaptation to OPAT warrants further investigation for the establishment of an optimal regimen.
In the last few years, three new intravenous alternatives have emerged. Firstly, daptomycin could be an alternative in this scenario. Among and within the studies assessing daptomycin efficacy in IE episodes, noteworthy heterogeneity was found in the therapeutic approach, but also in the conclusion reported. The poor results stated by Ceron et al.could dismiss the daptomycin recommendation supported by the promising results reported in other studies. Nevertheless, it should be noted that this study includes a small population and the data regarding patients switching between treatment groups are unclear.
Teicoplanin is another antibiotic possible alternative in this situation. It is a popular agent for OPAT owing to its long elimination half-life that enables once-daily dosing as well as intramuscular or subcutaneous administration, which could preclude permanent vascular access. The studies assessing the effectiveness of this alternative, which encompassed outpatient treatments, have shown promising outcomes. Nevertheless, the proposed dose regimen highly varies with respect to dose (5-10 mg/kg/day), and also loading dose use and length of therapy. As it currently stands, data providing support for the optimal dose and route of administration for teicoplanin options remain uncertain. Finally, Gram-positive IE treatment with dalbavancin has been studied, and included E. faecalis IE. However, the regimen design was inconsistent and the E. faecalis sample size small. Dalbavancin is a welcome agent in OPAT programs due to its extended dosing interval and the reduction of health-care visit needed, although it is not accessible worldwide due to its high cost. Nevertheless, for hospitals without OPAT programs, it might be a cost-effective alternative.
In conclusion, any of these alternatives should be considered for the continuation treatment of E. faecalis IE, but further research is needed to analyze the efficacy and safety in larger cohorts, and to standardize the optimal dose regimen. These two last options attract particular attention due to the inner benefits, previous experience in the outpatient use and management, and, in the case of teicoplanin, the number of successfully treated patients. Thus, they might be considered the two more reasonable intravenous alternatives for continuation treatment for E. faecalis IE patients.
Lastly, partial oral treatment for IE has been assessed. The authors demonstrated that switching to oral therapy was non-inferior to continued intravenous therapy after initial intravenous treatment. From their results, amoxicillin-based regimens emerge as a treatment option in this field. Nonetheless, oral therapy entails normal gastrointestinal uptake and good adherence, ensured in non-self-administered OPAT. Antibiotic exposure in high inoculum infections after oral antibiotic administration depends on the pharmacokinetics profile of each drug, and nowadays it is a matter of discussion, especially regarding β-lactam oral therapy. In this trial, seven patients in the oral therapy arm showed antibiotic plasma concentrations lower than the effective level, whereas no regimen adjustment was made on that base. Linezolid, an appealing option due to its bioavailability, has shown conflicting results, thus its use for IE treatment is not systematically recommended. Despite being an encouraging option, oral antibiotic therapy warrants further investigation to elucidate the best drug choice and regimen in each scenario, and also for patient selection.
The present review has a number of limitations and strengths. Observational studies were included due to the historical difficulties of performing randomized controlled trials in IE. Nonetheless, the study has been performed following PRISMA recommendation, and included an in-depth assessment of the studies' quality. For most of the alternatives discussed, inconsistent regimen designs and doses have been proposed. As such, in spite of the inclusion of a large enough population for drawing conclusions, in some of them, is not possible to recommend a precise regimen. Another limitation was the exclusion of studies based on the lack of information determined by microorganism or by treatment alternative, and the limited data available about actual OPAT IE treatment. The extrapolation of data from the inpatient to the outpatient setting could be inaccurate due to big differences in drug delivery and monitoring. Furthermore, incomplete retrieval is possible for papers outside the databases searched. Onn the other hand, we have conducted a rigorous search and systematic review accompanied by a narrative synthesis. Although previous work summarizing E. faecalis infective endocarditis treatment has been conducted lately, there is an absence of previous reviews gathering evidence concerning outpatient alternatives and possible adaptations of the current treatments, which highlights the novelty and relevance of this review. Our work focused on treatment alternatives suitable for outpatient treatment and early discharge options, which are of great clinical and scientific interest. Our analysis provides a sense of what alternatives could safely be used in this setting, and which alternatives are promising but require further research. It also provides a unique and valuable contribution to the available literature.
# Materials and methods
Our review protocol was prospectively registered on the PROSPERO international prospective register of systematic reviews (Protocol ID:154593) and is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.
## Search strategy
Our systematic search strategy was developed to seize all articles related to appropriate E. faecalis IE antimicrobial treatment in the outpatient setting. We conducted a search in the MEDLINE (through PubMed interface), EMBASE and Web of Science Core collection databases from inception until October 2019, using MeSH terms and keywords associated with the concepts presented in. The search strategy was amended according to the functionality of each of the databases. In addition, articles of interest identified by citation tracing were included.
Our systematic search strategy was developed to seize all articles related to appropriate E. faecalis IE antimicrobial treatment in the outpatient setting. We conducted a search in the MEDLINE (through PubMed interface), EMBASE and Web of Science Core collection databases from inception until October 2019, using MeSH terms and keywords associated with the concepts presented in. The search strategy was amended according to the functionality of each of the databases. In addition, articles of interest identified by citation tracing were included.
## Eligibility criteria
The inclusion of cohort studies and series of cases in a systematic review has been discouraged, however the lack of better evidence in the field forced us to include all types of human studies except case reports. Studies were eligible for review if they were published in English or Spanish and reported clinical outcomes from E. faecalis IE patients treated as inpatient or outpatient with oral or intravenous antibiotics appropriated for continuation therapy. We allowed the inclusion of studies reporting data from inpatient treatments where the antibiotics used were included in OPAT guidelines, or their use in OPAT was already reported, and considered them as "suitable for OPAT". Likewise, we included studies which partially included information not identified by microorganism or antibiotic regimen, but this data was dismissed. The eligibility criteria applied in this study are presented in detail in.
## Eligibility criteria
The inclusion of cohort studies and series of cases in a systematic review has been discouraged, however the lack of better evidence in the field forced us to include all types of human studies except case reports. Studies were eligible for review if they were published in English or Spanish and reported clinical outcomes from E. faecalis IE patients treated as inpatient or outpatient with oral or intravenous antibiotics appropriated for continuation therapy. We allowed the inclusion of studies reporting data from inpatient treatments where the antibiotics used were included in OPAT guidelines, or their use in OPAT was already reported, and considered them as "suitable for OPAT". Likewise, we included studies which partially included information not identified by microorganism or antibiotic regimen, but this data was dismissed. The eligibility criteria applied in this study are presented in detail in.
## Selection of studies
All the titles and abstracts of the citations identified by our database and manual search were screened after duplicate removal. Relevant articles or those with insufficient information within the Antibiotics 2020, 9, 657 13 of 17 title and abstract were full-text assessed. This selection was independently performed by two reviewers (LHH and MVGN) and, in case of uncertainty, discussed until consensus was reached.
## Quality assessment
To evaluate the quality of the studies selected for inclusion, we used three tools according to the study design: A Cochrane risk-of-bias tool for randomized trials (ROB.2) was used for randomized controlled trialsand the Risk of Bias in Non-Randomized Studies of Interventions tool (ROBINS-I) was suitable for non-randomized studies. For case series studies, quality was assessed using a standardized Study Quality Assessment Tool (SQAT) designed by the National Heart, Lung, and Blood Institute under the National Institutes of Health . When using the ROBINS-I tool the overall risk of bias of the paper was categorized into "Low", "Moderate", "Serious" or "Critical". When ROB.2 was applied, risk of bias was classified into the "Low", "High" or "Some concerns" categories. Finally, when using the NIH Quality Assessment Tool, the reported risk of bias was summarized as "Good", "Fair" or "Poor".
## Data extraction and synthesis
Data extraction was performed by one reviewer (LHH) using a standardized data extraction form; any uncertainty was discussed with another investigator (MVGN). We extracted the following information from the studies: Study design, treatment setting (outpatient or inpatient), type of endocarditis (left-or right-side endocarditis, native or prosthetic valve endocarditis or cardiac device-related endocarditis), endocarditis definition, follow-up period, treatment alternative (antibiotic or antibiotic combination, dose regimen and route of administration), number of E. faecalis IE patients treated, adverse events, clinical outcomes (mortality, relapses and others) and key findings. Given the heterogeneity of the study designs, interventions and outcome measures, it was not feasible to pool the results in a meta-analysis. Alternatively, we performed a narrative synthesis of evidence following the Cochrane Consumers and Communication Review Group's guidelines . Studies were grouped according to the antibiotic alternative reported, individual study characteristics and findings were summarized, and similarities, differences and patterns were identified.
# Conclusions
In this review, the therapeutic treatments against E. faecalis were reviewed, with special focus on outpatient therapy. The best option for continuation outpatient therapy after discharge is still unknown. The gold standard options for inpatient treatment require regimen adjustments which are poorly studied. New attractive alternatives for OPAT are arising, especially teicoplanin and dalbavancin regimens, the pharmacokinetics profiles and ease of administration of which provide significant advantages for outpatient treatment, whereas their safety and efficacy are not strongly evidence-supported yet. The assessment of the safety and efficacy of the suggested alternatives against E. faecalis IE in outpatient warrants future investigations. Funding: AGV was supported by the Instituto de Salud Carlos III, cofinanced by the European Development Regional Fund ("A way to achieve Europe"), Subprograma Miguel Servet (grant CP19/00159). LHH was supported by the Instituto de Salud Carlos III, Subprograma Rio Hortega (grant CM19/00152). |
trans-Acting Arginine Residues in the AAA+ Chaperone ClpB Allosterically Regulate the Activity through Inter- and Intradomain Communication*
Background: Two neighboring, trans-acting arginines in the N-terminal AAAϩ domain are essential for oligomerization and activity of ClpB/Hsp104. Results: Both arginines couple nucleotide binding to oligomerization and allosterically regulate the ATPase activity. Conclusion: Site-specifically engineered, cross-linked dimers of AAAϩ subunits can be utilized to study allosteric regulation. Significance: This study elucidates the mechanistic role of an essential arginine pair conserved in different AAAϩ proteins.The molecular chaperone ClpB/Hsp104, a member of the AAA؉ superfamily (ATPases associated with various cellular activities), rescues proteins from the aggregated state in collaboration with the DnaK/Hsp70 chaperone system. ClpB/Hsp104 forms a hexameric, ring-shaped complex that functions as a tightly regulated, ATP-powered molecular disaggregation machine. Highly conserved and essential arginine residues, often called arginine fingers, are located at the subunit interfaces of the complex, which also harbor the catalytic sites. Several AAA؉ proteins, including ClpB/Hsp104, possess a pair of such trans-acting arginines in the N-terminal nucleotide binding domain (NBD1), both of which were shown to be crucial for oligomerization and ATPase activity. Here, we present a mechanistic study elucidating the role of this conserved arginine pair. First, we found that the arginines couple nucleotide binding to oligomerization of NBD1, which is essential for the activity. Next, we designed a set of covalently linked, dimeric ClpB NBD1 variants, carrying single subunits deficient in either ATP binding or hydrolysis, to study allosteric regulation and intersubunit communication. Using this well defined environment of sitespecifically modified, cross-linked AAA؉ domains, we found that the conserved arginine pair mediates the cooperativity of ATP binding and hydrolysis in an allosteric fashion.ship from the German National Academic Foundation (to C. Z.).
The molecular disaggregation machine ClpB/Hsp104 (caseinolytic peptidase B/heat shock protein 104) is crucial for maintaining protein homeostasis because it reactivates aggregated proteins under cellular stress conditions in concert with the DnaK/Hsp70 chaperone system [bib_ref] Hsp104 and ClpB: protein disaggregating machines, Doyle [/bib_ref] [bib_ref] Mapping the road to recovery: the ClpB/Hsp104 molecular chaperone, Hodson [/bib_ref] [bib_ref] Aggregate reactivation mediated by the Hsp100 chaperones, Zolkiewski [/bib_ref]. Belonging to the superfamily of AAAϩ proteins (ATPases associated with various cellular activities), ClpB/Hsp104 functions as a hexameric complex that converts the chemical energy from ATP hydrolysis into mechanical force [bib_ref] The AAAϩ superfamily: a myriad of motions, Tucker [/bib_ref]. Protein disaggregation by ClpB/ Hsp104 involves the threading of single polypeptide chains out of the aggregate through the central pore of the hexameric ring [bib_ref] Thermotolerance requires refolding of aggregated proteins by substrate translocation through the central..., Weibezahn [/bib_ref]. High-resolution structural information is available for ClpB from Thermus thermophilus, showing a domain architecture that consists of a small N-terminal domain and two highly conserved AAAϩ domains, also called nucleotide binding domains (NBD1 2 and NBD2), per monomer [bib_ref] The structure of ClpB: a molecular chaperone that rescues proteins from an..., Lee [/bib_ref]. There is a long helical insertion into NBD1, named the M-domain, which was recently identified as a major regulatory element as well as the interaction site for the co-chaperone DnaK/Hsp70 .
The ATPase modules NBD1 and NBD2 are the motors that drive the molecular machine in a cooperative fashion. The catalytic sites, which are located at the interface between two subunits in the hexameric complex, are built up by highly conserved motifs, namely the Walker A and B motifs that are crucial for nucleotide binding and ATP hydrolysis, respectively [bib_ref] Distantly related sequences in the ␣-and -subunits of ATP synthase, myosin, kinases..., Walker [/bib_ref] [bib_ref] Structure and function of the AAAϩ nucleotide binding pocket, Wendler [/bib_ref]. Furthermore, there are essential arginine residues, often termed arginine fingers, that contribute to the active sites in trans because they are in close proximity to the nucleotide bound to the adjacent subunit. The role of such conserved arginines in AAAϩ proteins has been investigated extensively . However, it is not trivial to distinguish a truly catalytic arginine finger, as initially identified in GTPase-activating proteins [bib_ref] Confirmation of the arginine-finger hypothesis for the GAP-stimulated GTP-hydrolysis reaction of Ras, Ahmadian [/bib_ref] , from conserved arginines that either stabilize the hexameric state or are crucial for allosteric regulation. The complexity is even increased by the fact that several AAAϩ proteins, such as ClpB/Hsp104, ClpA, ClpC, and p97/VCP/ Cdc48, possess two highly conserved, neighboring arginines in their NBD1 subunit interface. With this study, we aimed at understanding the mechanistic role of this essential, trans-acting pair of arginines in allosteric communication between AAAϩ subunits.
We showed previously that NBD1-M and NBD2 of ClpB from T. thermophilus can be expressed and purified separately [bib_ref] Biochemical coupling of the two nucleotide binding domains of ClpB: covalent linkage..., Beinker [/bib_ref] , which allowed a detailed and quantitative characterization of both AAAϩ motor domains with regard to nucleotide bind-* This work was supported by the Max Planck Society and by a Ph.D. scholar-ing, oligomerization, and activity [bib_ref] Nucleotide Binding and Allosteric Modulation of the second AAAϩ domain of ClpB..., Werbeck [/bib_ref] [bib_ref] Coupling of oligomerization and nucleotide binding in the AAAϩ chaperone ClpB, Werbeck [/bib_ref] [bib_ref] Elements in nucleotide sensing and hydrolysis of the AAAϩ disaggregation machine ClpB:..., Zeymer [/bib_ref] [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. Here, we used the construct NBD1-M. Inspired by successful work on the mechanism of ClpX, another AAAϩ protein , we applied a combined approach of covalently linking NBD1-M subunits and introducing Walker A/B and arginine finger mutations. Using these fixed and well determined arrangements of wildtype and mutated subunits in a direct neighborhood, it was possible to dissect the mechanisms of allosteric regulation and intersubunit communication in the AAAϩ chaperone ClpB/Hsp104.
## Experimental procedures
Mutagenesis, Protein Expression, and Purification-The construct ClpB NBD1-M containing amino acids 141-534 of ClpB from T. thermophilus as well as the full-length protein ClpB from T. thermophilus were described previously [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref] [bib_ref] The chaperone function of ClpB from Thermus thermophilus depends on allosteric interactions..., Schlee [/bib_ref]. Sitedirected mutagenesis was performed using QuikChange PCR according to the QuikChange protocol (Agilent Technologies, Santa Clara, CA) and verified by DNA sequencing (Eurofins MWG Operon, Ebersberg, Germany). The mutations P221C, M394C, Q184C, A390C, R322A, R323A, K204Q, and E271Q were introduced as single mutations and/or in various pairwise combinations. Full-length ClpB and the truncated variant NBD1-M were expressed recombinantly in Escherichia coli BL21 (DE3) RIL and purified as described previously [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref] [bib_ref] The chaperone function of ClpB from Thermus thermophilus depends on allosteric interactions..., Schlee [/bib_ref]. 5 mM -mercaptoethanol was added to all buffers for the purification of cysteine-containing variants. Purified proteins were stored in buffer A (50 mM Tris/HCl, pH 7.5, 50 mM KCl, 5 mM MgCl 2 , and 2 mM EDTA), including 5 mM -mercaptoethanol for cysteine-containing variants.
Intermolecular Disulfide Bond Formation to Generate Covalently Linked ClpB Dimers-Cysteine residues were introduced to facilitate the formation of intermolecular disulfide bonds in the ClpB NBD1 subunit interface. The design was based on an available planar hexameric model of ClpB [bib_ref] Modeling AAAϩ ring complexes from monomeric structures, Diemand [/bib_ref]. Prior to the reaction, the reducing agent -mercaptoethanol was removed by buffer exchange. The formation of covalently linked dimers of full-length ClpB or NBD1-M variants using the single cysteine mutant pair P221C/M394C was performed in 50 mM Tris/ HCl, pH 7.5, 50 mM KCl, 5 mM MgCl 2 . The buffer was EDTAfree because 50 M copper phenanthroline was used as the oxidizing agent. Equimolar amounts of the respective cysteine variants (50 M each) were used in a 5-ml reaction volume. 2 mM ADP was added to trigger oligomerization of ClpB variants. The mixture was incubated for 1 h at 37°C. Subsequently, the reaction mixture was applied to a Superdex 200 26/60 size exclusion column equilibrated with buffer A to separate the formed dimer from unreacted monomer. The purity of the dimer products was evaluated by non-reducing SDS-PAGE. Test reactions were performed to ensure that homodimer formation was negligible. Another pair of cysteines (Q184C/ A390C) was also used to form a covalently linked dimer as described above. However, because this cross-linked variant showed severely impaired ATPase activity, it was not considered for further experiments.
Steady-state ATPase Measurements-Steady-state ATPase activity was measured in a coupled colorimetric assay at 25°C using a JASCO V-650 spectrophotometer (JASCO Germany GmbH, Gross-Umstatt, Germany). ClpB NBD1-M variants were incubated at 25°C in assay buffer (50 mM Tris/HCl, pH 7.5, 100 mM KCl, 2 mM EDTA, 0.4 mM phosphoenolpyruvate, 0.4 mM NADH, 0.1 g/liter BSA, 4 units/ml pyruvate kinase, 6 units/ml lactate dehydrogenase, and 10 mM MgCl 2 ). Importantly, reducing agents were strictly excluded from the assay buffer to maintain the intermolecular disulfide bonds of covalently linked dimers. The reaction was started by adding ATP (0.01-8 mM). Measurements were performed for protein concentrations of 1-30 M (with respect to monomeric units). The decreasing absorption at 340 nm was monitored over time, and the maximal slope was used to determine the ATPase turnover rate per monomer (⑀(NADH) ϭ 6220 M Ϫ1 cm Ϫ1 ). The data were analyzed with the Hill equation (Equation 1) using the program GraphPad Prism version 5.0.
[formula] k ϭ k cat ⅐ ͓ATP͔ n K m n ϩ ͓ATP͔ n (Eq. 1) [/formula]
Stopped Flow Experiments (Determination of MANT-dADP Binding Parameters)-Nucleotide binding experiments were performed with a BioLogic SFM-400 stopped flow instrument in single mixing configuration (BioLogic Science Instruments, Claix, France) in buffer A at 25°C essentially as described previously [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. The fluorescently labeled nucleotide MANT-dADP was purchased from BIOLOG (Bremen, Germany). The excitation wavelength was set to 296 nm, and the fluorescence signal was observed using a 400-nm long pass filter (400FG03-25, LOT Oriel Group). This setup was used to selectively excite protein-bound MANT-dADP via fluorescence resonance energy transfer (FRET) from the initially excited tryptophan residues of the protein. Kinetic traces were recorded as triplicates and averaged. Data analysis was performed using the program GraphPad Prism version 5.0.
Kinetic traces from direct binding experiments (2 M ClpB NBD1-M mixed 1:1 with 10 -50 M MANT-dADP) were fitted to exponential functions. The extracted rate constants were plotted against the nucleotide concentration. The association rate constants k on for MANT-dADP binding were obtained from the slope of the resulting linear functions. Kinetic traces from dissociation experiments (2 M ClpB NBD1-M incubated with 15 M MANT-dADP and subsequently mixed 1:1 with 5 mM Mg-ADP) were fitted to exponential functions; the extracted rate constants correspond to the dissociation rate constants k off of MANT-dADP binding. The K D was calculated from the ratio k off /k on . Given protein concentrations refer to monomeric units.
Fluorescence Equilibrium Titrations (Determination of ADP/ ATP Binding Parameters)-Fluorescence titrations were performed at 25°C in buffer A using a JASCO FP-8500 fluorescence spectrometer (JASCO Germany GmbH) as described previously [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. The excitation wavelength was set to 296 nm to facilitate selective excitation of protein-bound MANT-dADP via FRET from nearby tryptophan residues. The MANT fluorescence signal was monitored at 441 nm. Direct titrations of ClpB NBD1-M variants (at 2 or 20 M) with MANT-dADP (2-50 M) were used to determine the binding affinity of MANT-dADP, which was subsequently applied as the refer-ence K D in displacement titrations to determine K D (ADP) or K D (ATP). Here, ClpB NBD1-M variants (at 2 or 20 M) were incubated with MANT-dADP and subsequently titrated with ADP (2.5-300 M) or ATP (125-20,000 M). ATP titrations were performed in the presence of 2 mM phosphoenolpyruvate and 0.01 mg/ml pyruvate kinase (Roche Applied Science) as an ATP-regenerating system. The data were corrected for dilution effects and analyzed with a cubic equation for competing ligands using the initial concentrations of protein and MANT-dADP as well as the K D (MANT-dADP) from the direct titration as input values [bib_ref] Evaluation of human immunodeficiency virus type 1 reverse transcriptase primer tRNA binding..., Thrall [/bib_ref]. The program GraFit version 5.0 was used for data fitting.
Gel Filtration Experiments with Static Light Scattering (SLS) Analysis-Gel filtration experiments were performed on a Superdex 200 10/300 GL column connected to a refractive index detector (2414 from Waters (Milford, MA)), a photodiode array detector (2996 from Waters), and a multiangle light scattering detector (Dawn Heleos, Wyatt (Santa Barbara, CA)) in buffer A as described previously [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. The running buffer was either nucleotide-free or supplemented with 2 mM ADP or 2 mM ATP. 40 l of 100 M NBD1-M (with respect to monomeric units) were injected, resulting in a final concentration of about 2 M at the detector due to a 1:50 dilution by the gel filtration column. Molecular mass values were extracted from the multiangle light scattering data using the ASTRA software (Wyatt).
Dynamic Light Scattering (DLS) Experiments-DLS experiments were performed at protein concentrations of 1-30 M (with respect to monomeric units) in buffer A, which was either nucleotide-free or supplemented with 2 mM ADP or 2 mM ATP, respectively. In the case of ATP, 2 mM phosphoenolpyruvate and 0.01 mg/ml pyruvate kinase (Roche Applied Science) were present as an ATP-regenerating system. The measurements were performed with a Viscotek 802DAT DLS instrument (Viscotek, Waghäusel, Germany). 40 scans with a measuring time of 5 s/scan were recorded. Hydrodynamic radii and molecular mass values were extracted from the DLS data using the Omni-SIZE version 3.0 software package.
Disaggregation Assay (Chaperone-assisted Reactivation of Heat-aggregated ␣-Glucosidase)-The assay was performed to test whether disulfide cross-linking using the cysteine pair P221C/M394C affects the chaperone activity of full-length ClpB. 0.2 M ␣-glucosidase from Bacillus stearothermophilus was denatured for 8 min at 75°C in reaction buffer containing 50 mM MOPS, pH 7.5, 150 mM KCl, 10 mM MgCl 2 , and 5 mM ATP. Chaperones were added prior to refolding at 55°C. The total chaperone concentrations were c(ClpB)
[formula] ϭ 1.0 M, c(DnaK) ϭ 1.6 M, c(DnaJ) ϭ 0.5 M, and c(GrpE) ϭ 0.2 M. [/formula]
The co-chaperones DnaK, DnaJ, and GrpE from T. thermophilus were expressed and purified as described previously [bib_ref] The functional cycle and regulation of the Thermus thermophilus DnaK chaperone system, Klostermeier [/bib_ref]. Samples were taken after 30, 60, and 120 min and diluted 1:10 into assay buffer containing 50 mM KP i , pH 6.8, 2 mM paranitrophenyl-␣-D-glucopyranoside, 0.1 mg/ml BSA. The ␣-glucosidase activity was measured at 40°C using a microplate spectrophotometer (Varioskan, Thermo Electron, Vantaa, Finland). The average rate of absorption increase at 405 nm was monitored and normalized against a positive control containing ␣-glucosidase that was not heat-aggregated. Importantly, reducing agents were strictly excluded from all buffers to main-tain the intermolecular disulfide bonds of covalently linked ClpB dimers.
# Results
## Conserved arginines in clpb nbd1 mediate the coupling between nucleotide binding and oligomerization-a pair of conserved, neighboring arginines in clpb nbd1 (arg-322 and
Arg-323) is located at the interface between two subunits in the oligomeric ClpB complex [bib_ref] The structure of ClpB: a molecular chaperone that rescues proteins from an..., Lee [/bib_ref]. Both residues are in close proximity to the ATP molecule bound to the neighboring active site and were shown to be essential for the catalytic activity [bib_ref] Roles of conserved arginines in ATP-binding domains of AAAϩ chaperone ClpB from..., Yamasaki [/bib_ref]. In order to gain insights about the mechanistic role of these arginines, we decided to work with the truncated ClpB construct NBD1-M, which comprises only the N-terminal nucleotide binding domain (NBD1) and the helical M-domain of ClpB [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref]. We showed recently that this separately expressed construct is a fully active ATPase displaying a strong coupling between nucleotide binding and oligomerization as well as a highly cooperative ATPase activity, thereby reflecting important properties of full-length ClpB [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. We replaced both conserved arginine residues by alanine in ClpB NBD1-M. In agreement with previous experiments on full-length ClpB by , the single and double mutants R322A, R323A, and R322A/R323A, respectively, showed about 1000fold reduced ATPase activity compared with the wild type, indicating that both conserved arginines are crucial for ATP hydrolysis in ClpB NBD1. Next, we tested whether the mutations R322A and R323A influence the nucleotide binding behavior of ClpB NBD1-M. First, we performed stopped flow experiments to extract the kinetic nucleotide binding parameters for fluorescently labeled MANT-dADP [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref] and [fig_ref] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A,... [/fig_ref]. Direct mixing and dissociation experiments showed that both the single and double mutants were able to bind MANT-dADP with similar affinities compared with the wild type. This result is in agreement with previous studies on full-length ClpB [bib_ref] Roles of conserved arginines in ATP-binding domains of AAAϩ chaperone ClpB from..., Yamasaki [/bib_ref]. Furthermore, we determined the binding affinities for the unlabeled nucleotides ADP and ATP by displacement titrations at low and high protein concentration (2 and 20 M, respectively) [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref] and [fig_ref] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A,... [/fig_ref]. Notably, the nucleotide binding affinities of the single mutants R322A and R323A and the double mutant R322A/R323A did not increase at higher protein concentrations as observed for the wild type, indicating that the conserved arginines are involved in coupling nucleotide binding and oligomerization. To further substantiate this hypothesis, we characterized the oligomerization behavior upon nucleotide binding for the single and double mutants using both DLS and SLS experiments [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref] and [fig_ref] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A,... [/fig_ref]. In the presence of ATP, the wild-type protein NBD1-M oligomerizes. With increasing protein concentration, a shift toward trimeric species is observed, which correlates tightly with the increase in ATP hydrolysis rates [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref]. This together with Hill coefficients higher than 2.5 indicates that the trimer represents the smallest hydrolysis-competent unit [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref]. The nucleotideinduced oligomerization of NBD1-M is severely impaired by the R322A and R323A mutations. In contrast to the observed trimers for the wild-type protein, the molecular masses obtained for the single and double mutants indicate only a monomer/dimer equilibrium [fig_ref] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and... [/fig_ref] and [fig_ref] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A,... [/fig_ref]. Due to sub- Single exponential fits are shown as colored lines. Similar traces were obtained for the single mutants R322A and R323A. The rate constants k obs obtained from fitting the kinetic traces are plotted against the MANT-dADP concentration (right). The association rate constants k on for MANT-dADP binding were obtained from the slope of the linear functions. The dissociation rate constants k off can be estimated from the y axis intercepts, but they were determined separately in dissociation experiments, as described under "Experimental Procedures." Nucleotide binding parameters extracted from these data are listed in [fig_ref] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A,... [/fig_ref]. C, fluorescence equilibrium titrations. NBD1-M was incubated with MANT-dADP and subsequently titrated with ADP or ATP, respectively. For the ATP titration, phosphoenolpyruvate and pyruvate kinase were present as an ATP-regenerating system. The volumecorrected data were fitted with the cubic equation for competing ligands (31), using K D (MANT-dADP) as an input value. This figure shows the titrations for NBD1-M R322A; similar curves were obtained for all mutants at different protein concentrations. Nucleotide binding parameters extracted from these data are listed in [fig_ref] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A,... [/fig_ref]. D, analytical gel filtration with SLS analysis. Elution profiles of NBD1-M in nucleotide-free buffer (red) and with 2 mM ADP (blue) and 2 mM ATP (green) present in the running buffer are shown for NBD1-M wild type (top) and NBD1-M R322A/R323A (bottom). The ATP-containing buffer was supplemented with phosphoenolpyruvate and pyruvate kinase as an ATP-regenerating system. Solid line, refractive index signal; dotted line, calculated molecular mass of the eluted species. The actual molecular mass of the NBD1-M monomer is 45 kDa. E, correlation between oligomeric state and ATPase activity. The steady-state ATPase turnover (orange circles) and the molecular mass of oligomeric NBD1-M species (green bars) measured by DLS are plotted in the same diagram for different protein concentrations. The DLS experiments were performed in the presence of 2 mM ATP and phosphoenolpyruvate and pyruvate kinase as an ATP-regenerating system. The increase in ATPase activity correlates with the formation of NBD1-M trimers (F), which represent the smallest ATP hydrolysis-competent unit. a.u., arbitrary units. stantial dilution during gel filtration, the molecular masses obtained from SLS data are not as high as in the DLS measurements performed at about 10-fold higher protein concentration. Still, the nucleotide-induced shift of the elution peak is suppressed for the mutated variants, and the obtained masses are significantly lower. It can be concluded that, although the conserved arginines Arg-322 and Arg-323 are not essential for nucleotide binding competence, they mediate the coupling between nucleotide binding and oligomerization. They are key structural elements required for the nucleotide-induced oligomerization of ClpB, a prerequisite for activity.
## Covalently linked clpb dimers facilitate mechanistic studies on allosteric
Regulation-With the intention to generate a well defined and fixed ClpB subunit interface, we designed covalently linked NBD1-M dimers with an intermolecular disulfide cross-link [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. Two pairs of cysteines were tested, namely P221C/M394C and Q184C/A390C. The positions were chosen on the basis of an available planar, hexameric ClpB model [bib_ref] Modeling AAAϩ ring complexes from monomeric structures, Diemand [/bib_ref] using the program SSBOND, which suggests positions for cysteine pairs according to the optimal distances and dihedral angles for disulfide bonds [bib_ref] Model building of disulfide bonds in proteins with known three-dimensional structure, Hazes [/bib_ref]. Whereas the Q184C/ A390C dimer was severely impaired in ATPase activity (100- [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. The parameters are given here for comparison. b From stopped flow experiments. c From fluorescence equilibrium titrations. d Measurements were performed in the absence and presence of nucleotide. The molecular mass of monomeric NBD1-M is 45 kDa. e ATP-containing solutions were supplemented with phosphoenolpyruvate and pyruvate kinase as an ATP-regenerating system. fold lower than the unlinked wild-type, data not shown), the P221C/M394C dimer showed ATP hydrolysis rates comparable with the wild-type. Furthermore, we tested whether the intermolecular disulfide bond between P221C and M394C affects the chaperone activity of ClpB. To this end, we generated an identically cross-linked variant of the full-length protein, which was active in an ␣-glucosidase disaggregation assay with about 70% activity of the unlinked wild-type protein [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. Thus, the P221C/M394C disulfide cross-link was considered "minimally invasive" and was used for all further experiments.
First, we determined the nucleotide binding parameters of the cross-linked NBD1-M dimer. Stopped flow experiments showed a biphasic fluorescence signal change upon direct mixing with MANT-dADP [fig_ref] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. Both kinetic phases were nucleotide concentration-dependent, indicating the presence of an asymmetric dimer with two unequal nucleotide binding sites. In order to assign the observed phases, we utilized NBD1-M dimers carrying the Walker A mutation K204Q in one of the two subunits [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. These dimers are deficient in nucleotide binding either in the cross-linked or the free active site. Indeed, MANT-dADP binding was monophasic for both variants but with significantly different kinetic parameters (k on , k off , and K D ), which allowed an unambiguous assignment [fig_ref] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig [/fig_ref]. Notably, K D (MANT-dADP) is significantly lower compared with the unlinked NBD1-M wild type for both active sites of the cross-linked dimer, mainly due to a decrease in the dissociation rate constant k off . This finding again confirms the strong coupling between nucleotide binding and oligomerization in ClpB NBD1. Both association (k on ) and dissociation (k off ) of nucleotide are slower for the cross-linked active site than for the free one. We furthermore determined the binding affinities for unlabeled ADP and ATP by fluorescence displacement titrations and observed 10-fold stronger ADP and 18-fold stronger ATP binding compared with the unlinked NBD1-M [fig_ref] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig [/fig_ref].
Next, we characterized the steady-state ATPase activity of the cross-linked NBD-M dimer [fig_ref] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 3: ATPase activity parameters of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. In agreement with the observed improvement in ATP binding, the dimer showed a significantly lower K m and less pronounced dependence of the activity on protein concentration compared with the unlinked NBD1-M. However, the maximum k cat observed for high protein concentrations of unlinked wild-type protein was not reached by the cross-linked dimer. It can be speculated that this is due to the fact that ADP release becomes rate-limiting, considering the measured k off (MANT-dADP) is lower than 0.1 s Ϫ1 for the cross-linked active site. We next generated NBD1-M dimers carrying the Walker B mutation E271Q in one of the two subunits [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. Using these dimers, which are fully nucleotide binding-competent but deficient in ATP hydrolysis either in the cross-linked or the free active site, we showed that indeed both of the two different active sites contribute to the overall activity of the dimer [fig_ref] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 3: ATPase activity parameters of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. When having the Walker B mutation in the free active site, the remaining activity originating from the crosslinked site is 25% of the cross-linked wild-type, whereas when mutating the cross-linked active site, the free active site is even 30% more active than the wild-type dimer carrying two intact subunits. This somewhat unexpected result emphasizes the importance of allosteric regulation. It seems that a tightly bound ATP molecule in the cross-linked active site activates the neighboring, free active site. To further study this phenomenon, we next measured the ATPase activity of the NBD1-M dimers carrying the Walker A mutation K204Q in one of the two subunits [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref] that were already used to assign the nucleotide binding phases. Notably, both variants were inactive [fig_ref] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. Independent of whether the cross-linked or free active site was mutated, it was sufficient to provide a nucleotide binding-deficient nearest neighbor to totally abolish the activity of the intact active site.
Furthermore, it was important to characterize the nucleotide-induced oligomerization of the cross-linked NBD1-M dimer variants. We performed analytical gel filtration runs with SLS analysis [fig_ref] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. The cross-linked wild-type dimer as well as both variants with the Walker B mutation showed pronounced ATP-induced oligomerization. A clear shift toward The obtained rate constants k obs plotted against the MANT-dADP concentration result in two linear functions (circles and triangles). In order to assign the respective association rate constants k on and dissociation rate constants k off to the two different binding sites, cross-linked NBD1-M dimers with Walker A mutations (see [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref] were used, which showed only monophasic kinetic traces upon direct mixing with MANT-dADP. Nucleotide binding parameters extracted from these data are listed in [fig_ref] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig [/fig_ref]. B, fluorescence equilibrium titrations. Cross-linked NBD1-M dimer variants were incubated with MANT-dADP and subsequently titrated with ADP or ATP, respectively (wild type (circles), R322A (squares), and R323A (triangles)). For the ATP titrations, phosphoenolpyruvate and pyruvate kinase were present as an ATP-regenerating system. The volume-corrected data were fitted with the cubic equation for competing ligands (31), using K D (MANT-dADP) as an input value. Nucleotide binding parameters extracted from these data are listed in [fig_ref] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig [/fig_ref]. C, analytical gel filtration with SLS analysis. Elution profiles of cross-linked NBD1-M dimers in nucleotide-free buffer (red) and with 2 mM ADP (blue) and 2 mM ATP (green) present in the running buffer are shown. Top, nucleotide-induced association of cross-linked, dimeric species was observed for wild type, Walker B mutants, and mutants of the conserved arginines (only one representative data set is shown). Bottom, mutants carrying a Walker A mutation do not form higher oligomers (only one representative data set is shown). The different cross-linked dimer variants are illustrated schematically, as introduced in [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. The ATP-containing buffer was supplemented with phosphoenolpyruvate and pyruvate kinase as an ATP-regenerating system. The refractive index signal is shown as a solid line, and the calculated molecular mass of the eluted species is shown as a dotted line. The actual molecular mass of the NBD1-M monomer is 45 kDa. a.u., arbitrary units. the size of tetramers was observed, indicating that also the cross-linked dimers associate with each other to form higher oligomers in the presence of ATP. In contrast, for both dimer variants carrying the Walker A mutation no nucleotide-induced oligomerization was observed, which is especially interesting for the variant with the cross-linked site being mutated. Here, the absence of nucleotide must be somehow communicated such that the intact and nucleotide binding-competent neighboring subunit cannot associate with another molecule. In summary, the cross-linked NBD1-M dimers allowed a dissection of allosteric effects in ClpB by site-specifically introducing Walker A and B mutations.
Allosteric Regulation between ClpB Subunits Is Communicated by Conserved Arginines-As a next step, it would be desirable to understand the molecular basis of how the observed allosteric regulation is communicated throughout the ClpB oligomer and whether the conserved arginines, Arg-322 and Arg-323, located in the subunit interfaced are involved in this task. To this end, we studied covalently linked NBD1-M dimer variants carrying either the R322A or R323A mutation in the cross-linked interface [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. This approach was chosen to distinguish a truly regulatory function of the arginines from effects associated with oligomeric stability, the latter of which were assumed to be blanked by covalently fixing the subunit interface. Indeed, SLS measurements confirmed that nucleotide-induced oligomerization of the cross-linked dimers was not impaired by the arginine to alanine mutations [fig_ref] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. In line with this result, both crosslinked dimer variants (R322A and R323A) were nucleotide binding-competent. The biphasic kinetic traces observed upon mixing with the fluorescently labeled MANT-dADP indicated the presence of two intact nucleotide binding sites per dimer. However, slightly higher K D values for MANT-dADP were obtained for the mutants compared with the cross-linked wild-type dimer [fig_ref] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig [/fig_ref]. In fluorescence displacement titrations, ADP binding was essentially not affected by the arginine to alanine mutations in the cross-linked interface, whereas ATP binding was significantly impaired [fig_ref] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig [/fig_ref]. This result suggests a The terms "free" and "cross-linked" refer to the two different nucleotide binding sites, one being located in the free (outside) interface and the other one in the cross-linked (inside) interface.
that Arg-322 and Arg-323 interact primarily with the ␥-phosphate group of the ATP molecule bound to the neighboring subunit, presumably sensing the nucleotide binding state that way. Next, we measured the steady-state ATPase activity of the NBD1-M dimers carrying the R322A or R323A mutation in the cross-linked interface [fig_ref] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 3: ATPase activity parameters of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. For both variants, the obtained K m values were significantly increased compared with the cross-linked wild-type dimer, indicating again that the conserved arginines contribute essential interactions that generate cooperativity. Notably, the R323A mutation caused a more severe loss in activity than the R322A mutation (90 and 35% decreased k cat compared with the wild-type dimer, respectively), which might indicate that Arg-322 is mainly involved in stabilizing the subunit interface, which was provided here by the disulfide cross-link. We next asked whether the conserved arginines indeed regulate the activity of a ClpB oligomer by allosteric communication, thus affecting a neigh-boring catalytic site that they do not directly interact with. To this end, we combined the R322A and R323A mutation with the Walker B mutation E271Q in the cross-linked interface, which allowed studying the direct effect of the arginine mutation on the activity of the catalytic site located in the neighboring, free interface [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref]. Although fully competent in ATP-induced oligomerization, these dimers showed a severely impaired steady-state ATPase activity compared with the dimer carrying only the Walker B mutation [fig_ref] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants [/fig_ref] and [fig_ref] TABLE 3: ATPase activity parameters of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. The cooperativity seemed to be totally blocked (Hill coefficient n ϭ 1), and very high K m , low k cat , and a strong dependence on protein concentration were observed. These results clearly confirm that the conserved arginines Arg-322 and Arg-323 regulate the highly cooperative ATPase activity of ClpB by allosterically communicating between neighboring subunits, which exceeds the simple mediation of ATP-induced oligomerization. . The corresponding kinetic parameters K m , k cat , and Hill coefficient n are listed in [fig_ref] TABLE 3: ATPase activity parameters of covalently linked ClpB NBD1-M dimer variants [/fig_ref]. The key at the bottom uses the schematic representations of the different cross-linked dimer variants as introduced in [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref] as follows: gray circles, unlinked NBD1-M wild type; green circles, cross-linked NBD1-M dimer wild-type; red circles, cross-linked NBD1-M dimer with Walker A mutation K204Q in the cross-linked interface; red triangles, cross-linked NBD1-M dimer with Walker A mutation K204Q in the free interface; orange circles, cross-linked NBD1-M dimer with Walker B mutation E271Q in the cross-linked interface; orange triangles, cross-linked NBD1-M dimer with Walker B mutation E271Q in the free interface; blue circles, cross-linked NBD1-M dimer with R322A mutation in the cross-linked interface; pink circles, cross-linked NBD1-M dimer with R323A mutation in the cross-linked interface; blue squares, cross-linked NBD1-M dimer with R322A mutation and Walker B mutation E271Q in the cross-linked interface; pink squares, cross-linked NBD1-M dimer with R323A mutation and Walker B mutation E271Q in the cross-linked interface.
# Discussion
In this study, we investigated the role of the conserved, transacting arginines Arg-322 and Arg-323 in allosteric regulation and intersubunit communication in the molecular disaggregation machine ClpB . Using a simplified system, namely the separate N-terminal ATPase subunit NBD1-M, it was possible to study the interplay between nucleotide binding, oligomerization, and activity in a quantitative manner. We utilized a set of well defined NBD1-M dimers with intermolecular disulfide cross-links and site-specifically introduced Walker A/B mutations to draw conclusions about allosteric effects mediated by the conserved arginine pair.
First, we showed that both arginines are involved in coupling nucleotide binding to oligomerization of ClpB NBD1-M. We identified the NBD1-M trimer as the smallest ATP hydrolysiscompetent unit, which is formed upon ATP binding, but only if both arginines are present. The finding that trimer formation is essential is in agreement with previously performed mixing experiments showing that the random incorporation of two mutant ClpB subunits into the hexamer is sufficient to abolish activity [bib_ref] Coupling and dynamics of subunits in the hexameric AAAϩ chaperone ClpB, Werbeck [/bib_ref]. A previous study using full-length ClpB came to the conclusion that the trans-acting arginines are not involved in nucleotide binding [bib_ref] Roles of conserved arginines in ATP-binding domains of AAAϩ chaperone ClpB from..., Yamasaki [/bib_ref]. However, using the cross-linked dimers, we showed that the arginines are indeed crucial for strong and cooperative ATP binding.
The next goal was to obtain a better understanding of allosteric regulation mechanisms implemented in ClpB NBD1-M. A comprehensive mechanistic interpretation of the nucleotide binding, oligomerization, and activity data that we obtained for the different cross-linked dimer variants would greatly benefit from additional structural information about the ClpB subunit interface. The available crystal structure of ClpB from T. thermophilus (Protein Data Bank code 1QVR) exhibits a helical arrangement of subunits, thereby displaying a shifted subunit interface [bib_ref] The structure of ClpB: a molecular chaperone that rescues proteins from an..., Lee [/bib_ref]. The two conserved arginines Arg-322 and Arg-323 are located 5 and 11 Å away from the ␥-phosphate of ATP bound to the neighboring subunit, respectively (see [fig_ref] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds [/fig_ref] , which may not reflect the active conformation. Several cryo-EM studies on ClpB and its yeast homolog Hsp104 generated models of a planar hexameric ring, which is believed to be the active form [bib_ref] Visualizing the ATPase cycle in a protein disaggregating machine: structural basis for..., Lee [/bib_ref] [bib_ref] CryoEM structure of Hsp104 and its mechanistic implication for protein disaggregation, Lee [/bib_ref]. However, structural details, such as the conformation of the conserved arginines, could not be resolved. When using the hexameric crystal structure of the highly homologous AAAϩ protein ClpC together with its adaptor protein MecA (Protein Data Bank code 3PXG) as a template for a planar ClpB model, both conserved arginines are at a 4 -6-Å distance from a modeled ATP molecule [bib_ref] Structure and mechanism of the hexameric MecA-ClpC molecular machine, Wang [/bib_ref]. Still, at this point, there is no reliable knowledge about the exact positioning of the conserved arginine pair in the ClpB subunit interface. Thus, we put great emphasis on control experiments using different Walker A/B mutants to verify our results. [fig_ref] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants [/fig_ref] , A, C, and E, for corresponding experimental data. The ATPase activity was calculated per monomer. The Hill equation was used for data fitting (Equation 1).
a The characterization of ClpB NBD1-M wild-type has been published previously [bib_ref] The molecular mechanism of Hsp100 chaperone inhibition by the prion curing agent..., Zeymer [/bib_ref]. The parameters are given here for comparison. b The data showed no sigmoidal shape and were therefore fitted using the Michaelis-Menten equation, which is represented by a simple hyperbolar function (Hill coefficient n ϭ 1).
## Figure 5. allosteric regulation and intersubunit communication in clpb
NBD1. Putative pathways of allosteric regulation (orange arrows) are illustrated schematically. The N-terminal ATPase subunit of ClpB, NBD1, is shown as a gray sphere. ATP is depicted in green and the trans-acting arginines Arg-322 and Arg-323 are shown in blue and pink, respectively. The active site is located at the interface between two subunits in the oligomeric ClpB complex. The conserved arginines interact with the ␥-phosphate of the ATP molecule bound to the neighboring subunit, which presumably is the structural basis for intersubunit communication. The nucleotide state of the adjacent active site is sensed and communicated throughout the AAAϩ domain, which allosterically regulates the activity of the whole oligomeric complex. This process may involve other highly conserved residues that are part of nucleotide sensor motifs in cis and induce conformational changes throughout the ATPase cycle.
[fig] FIGURE 1: Characterization of ClpB NBD1-M wild type, single mutants R322A and R323A, and double mutant R322A/R323A. A, the ClpB construct NBD1-M comprises amino acids 141-534, which are highlighted in dark gray in the crystal structure of full-length ClpB (Protein Data Bank code 1QVR). The schematic representation for this construct is a gray sphere. ATP is shown in green. The conserved arginines Arg-322 and Arg-323 are shown in blue and pink, respectively. B, stopped flow nucleotide binding experiments. Kinetic fluorescence traces upon direct mixing of NBD1-M R322A/R323A and MANT-dADP are shown (left). The final concentration of protein is 1 M in all cases. The final MANT-dADP concentration ranges from 5 to 25 M. [/fig]
[fig] FIGURE 2: Covalently linked ClpB NBD1-M dimer variants with intermolecular disulfide bonds.A, schematic representation of a disulfide-linked NBD1-M dimer. The structural close-up view is based on the available crystal structure of ClpB (Protein Data Bank code 1QVR) and shows the catalytic site located in the interface between two NBD1 subunits (green and yellow, respectively) with the conserved, trans-acting arginines Arg-322 and Arg-323 in close proximity to the ATP molecule bound to the neighboring subunit. Lys-204 and Glu-271 are part of the catalytically essential Walker A and Walker B motifs, respectively. Two pairs of cysteines, P221C/M394C and Q184C/A390C, were introduced by site-directed mutagenesis to form intermolecular disulfide bonds, but only the first pair was used in further experiments. Different mutated dimer variants were designed, as shown in C-F. B, disaggregation activity of a full-length ClpB dimer cross-linked by the P221C/M394C disulfide bond. The relative ␣-glucosidase activity (normalized against the positive control) for different time points during the ClpB/DnaK/DnaJ/GrpE-assisted disaggregation reaction is shown for the unlinked ClpB wild-type protein (red) and the cross-linked ClpB dimer (blue). Negative controls were as follows: no chaperones present (black), only ClpB present (wild type (gray) and cross-linked dimer (yellow)), and only DnaK/DnaJ/ GrpE present (brown). C, NBD1-M dimers with Walker A mutation K204Q either in the cross-linked interface or in the free interface. The mutation leads to a nucleotide binding-deficient catalytic site. D, NBD1-M dimers with Walker B mutation E271Q either in the cross-linked interface or in the free interface. The mutation leads to an ATP hydrolysis-deficient catalytic site that remains nucleotide binding-competent. E, NBD1-M dimers with R322A or R323A mutation in the cross-linked interface. F, NBD1-M dimers with R322A or R323A mutation in combination with Walker B mutation E271Q in the cross-linked interface. [/fig]
[fig] FIGURE 3: Nucleotide binding and oligomerization of covalently linked ClpB NBD1-M dimer variants. A, stopped flow nucleotide binding experiments. Direct mixing of cross-linked wild-type NBD1-M dimer and MANT-dADP leads to biphasic kinetic traces. Double exponential fits are shown as colored lines. [/fig]
[fig] FIGURE 4: ATPase activity of covalently linked ClpB NBD1-M dimer variants. Steady-state ATPase turnover rates per monomer were measured at different ATP concentrations (A, C, and E) and protein concentrations (B, [/fig]
[table] TABLE 1: Nucleotide binding and oligomerization parameters of ClpB NBD1-M wild type, R322A, R323A, and R322A/R323A a The characterization of ClpB NBD1-M wild type has been published previously [/table]
[table] TABLE 2: Nucleotide binding parameters of covalently linked ClpB NBD1-M dimer variants See Fig. 3 for corresponding experimental data. Stopped flow experiments were performed at [NBD1-M] ϭ 1 M in the final mixture. Equilibrium titrations were performed at [NBD1-M] ϭ 2 M. n.d., not determined. [/table]
[table] TABLE 3: ATPase activity parameters of covalently linked ClpB NBD1-M dimer variants [/table]
|
Effects of Auditory Stimuli in the Horizontal Plane on Audiovisual Integration: An Event-Related Potential Study
This article aims to investigate whether auditory stimuli in the horizontal plane, particularly originating from behind the participant, affect audiovisual integration by using behavioral and event-related potential (ERP) measurements. In this study, visual stimuli were presented directly in front of the participants, auditory stimuli were presented at one location in an equidistant horizontal plane at the front (0u, the fixation point), right (90u), back (180u), or left (270u) of the participants, and audiovisual stimuli that include both visual stimuli and auditory stimuli originating from one of the four locations were simultaneously presented. These stimuli were presented randomly with equal probability; during this time, participants were asked to attend to the visual stimulus and respond promptly only to visual target stimuli (a unimodal visual target stimulus and the visual target of the audiovisual stimulus). A significant facilitation of reaction times and hit rates was obtained following audiovisual stimulation, irrespective of whether the auditory stimuli were presented in the front or back of the participant. However, no significant interactions were found between visual stimuli and auditory stimuli from the right or left. Two main ERP components related to audiovisual integration were found: first, auditory stimuli from the front location produced an ERP reaction over the right temporal area and right occipital area at approximately 160-200 milliseconds; second, auditory stimuli from the back produced a reaction over the parietal and occipital areas at approximately 360-400 milliseconds. Our results confirmed that audiovisual integration was also elicited, even though auditory stimuli were presented behind the participant, but no integration occurred when auditory stimuli were presented in the right or left spaces, suggesting that the human brain might be particularly sensitive to information received from behind than both sides.
# Introduction
In a social environment, many objects and events are often perceived simultaneously via different sensory systems. The information received from the different modalities must be localized and integrated by several systems to produce coherent cognition in the brain. Previous studies have shown that bimodal audiovisual stimuli can be discriminated or detected more accurately and faster than unimodal auditory or visual stimuli presented alone [bib_ref] Multisensory auditory-visual interactions during early sensory processing in humans: a high-density electrical..., Molholm [/bib_ref] [bib_ref] An analysis of audio-visual crossmodal integration by means of event-related potential (ERP)..., Teder-Sä Lejä Rvi [/bib_ref] [bib_ref] Audiovisual interaction enhances auditory detection in late stage: an event-related potential study, Li [/bib_ref] [bib_ref] Divided attention: Evidence for coactivation with redundant signals, Miller [/bib_ref]. This facilitative effect is called ''audiovisual integration''. In previous studies, discrimination tasks were used to investigate audiovisual integration using target and standard stimuli, and subjects were required to respond only to target stimuli [bib_ref] An analysis of audio-visual crossmodal integration by means of event-related potential (ERP)..., Teder-Sä Lejä Rvi [/bib_ref] [bib_ref] Audiovisual interaction enhances auditory detection in late stage: an event-related potential study, Li [/bib_ref]. Specifically, discrimination tasks involve visual (or auditory) detection methods in which subjects are instructed to respond to visual (or auditory) target stimuli while not responding to standard stimuli [bib_ref] Audiovisual interaction enhances auditory detection in late stage: an event-related potential study, Li [/bib_ref] [bib_ref] Involuntary orienting to sound improves visual perception, Mcdonald [/bib_ref] or visual and auditory detection methods in which subjects are required to respond to both visual and auditory target stimuli [bib_ref] Early auditory-visual interactions in human cortex during nonredundant target identification, Fort [/bib_ref] [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref].
In behavioral studies, using either visual and auditory detection [bib_ref] Spatial disparity affects visual-auditory interactions in human sensorimotor processing, Harrington [/bib_ref] [bib_ref] Timecourse of coactivation in bimodal divided attention, Miller [/bib_ref] or one modality (i.e., visual stimuli) detection [bib_ref] Enhancement of visual perception by crossmodal visuo-auditory interaction, Frassinetti [/bib_ref] , some researchers have shown that the facilitative effect of bimodal stimuli on signal detectability depend on the spatial location of the stimuli and that the modalities in the same location can produce the greatest response enhancements, whereas spatially disparate modalities induce a less potent effect or no change in response [bib_ref] The Merging of the Senses, Stein [/bib_ref]. Moreover, previous studies based on visual discrimination task have also confirmed that the perception of visual stimuli can be improved by simultaneous auditory stimuli occurring in close spatial location to the visual stimuli [bib_ref] Improvement of visual contrast detection by a simultaneous sound, Lippert [/bib_ref] [bib_ref] Crossmodal object-based attention: Auditory objects affect visual processing, Turatto [/bib_ref]. Recently, some researchers investigated audiovisual integration elicited by a pair of visual and auditory stimuli presented in the frontal horizontal plane (0u from the fixation point) and right locations (up to 90ufrom the fixation point) [bib_ref] Interactions between the spatial and temporal stimulus factors that influence multisensory integration..., Stevenson [/bib_ref]. Their findings indicated that the facilitative effects in localizing paired visual and auditory stimuli decrease from the central to peripheral locations and that this enhancement was no longer observed for the 90u right location. However, Stein, London, Wilkinson, and Price (1996) used a visual discrimination task to determine that the perception of peripheral visual stimuli (30u to the left or right of fixation) were unaffected by an auditory stimulus that was presented at the same location as the visual stimuli or at 45u to the right or left of the visual stimuli. When the visual stimuli were presented at the central fixation point, perception and detection of the visual stimuli were enhanced by simultaneous auditory stimuli, regardless of whether the auditory stimulus was spatially congruent or displaced 45u to the right or left of fixation [bib_ref] Enhancement of Perceived Visual Intensity by Auditory Stimuli: A Psychophysical Analysis, Stein [/bib_ref]. Therefore, the positional relationship of modalities was an unimportant factor in determining whether multisensory interactions had occurred.
In some event-related potential (ERP) studies, audiovisual (AV) integration was investigated by comparing the ERP from a bimodal AV stimulus with the sum of the ERPs of the constituent auditory (A) and visual (V) stimuli [bib_ref] Visual speech speeds up the neural processing of auditory speech, Van Wassenhove [/bib_ref] [bib_ref] Bimodal speech: early suppressive visual effects in human auditory cortex, Besle [/bib_ref] [bib_ref] Interest and validity of the additive model in electrophysiological studies of multisensory..., Besle [/bib_ref] [bib_ref] Spatial and temporal factors determine auditory-visual interactions in human saccadic eye movements, Frens [/bib_ref]. In these studies, the formula ERP (AV) -[ERP (A)+ERP (V)] = ERP (A 6 V interactions) was used to estimate cross-modal interaction or integration in the differences between the brain responses to bimodal stimuli and the algebraic sum of the unimodal responses. Moreover, the [AV-(A+V)] complex was used to determine cortical regions that were uniquely activated by bimodal stimulation. The effects of AV integration were expected to be observed as differences between the AV and the (A+V) waveforms [AV-(A+V)]. Based on this method, [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref] reported that integrative processes can occur as latencies as early as 40 milliseconds (ms) after stimulus onset over the posterior scalp areas when visual and auditory stimuli are simultaneously presented in the central location using a discrimination task [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref].
Furthermore, audiovisual integration of a spatial pair of visual and auditory stimuli was investigated using visual and auditory discrimination task with ERP recordings [bib_ref] Effects of Spatial Congruity on Audio-Visual Multimodal Integration, Teder-Sä Lejä Rvi [/bib_ref] [bib_ref] Multisensory processing in the redundant-target effect: A behavioral and event-related potential study, Gondan [/bib_ref]. In studies by Teder-Sä lejä rvi et al. , a pair of auditory and visual stimuli were presented either at the same spatial location (i.e., 30u to the left or right of the central fixation point) or at opposite spatial locations (i.e., V left and A right). Their findings showed that responses were faster for bimodal stimuli than for unimodal stimuli regardless of whether the stimuli were in the same or opposite spatial locations. However, [bib_ref] Multisensory processing in the redundant-target effect: A behavioral and event-related potential study, Gondan [/bib_ref] found that the differential integration between the same spatial location and the opposite spatial location was observed as early as 160 ms after stimulus onset over the parietal cortex using ERPs. Therefore, studies using high-density ERP recordings have revealed important insights about when (i.e., latency) and where (i.e., topography) audiovisual integration occurs in the human brain and indicated that the effect of integration was earlier in central, as opposed to peripheral, locations. These electrophysiological findings also confirmed that one modality can be affected by another modality, even when these modalities occur from incongruent spatial locations.
In these previous studies, audiovisual integration was studied using auditory and visual stimuli in various locations; however, these modalities were presented either directly in front of the participants, or to their left or right. Some recent studies have investigated audiovisual speech perception in which the range of locations for auditory stimuli and visual stimuli are extended to include those behind the participants (180u from the point of fixation) (i.e., the entire horizontal plane) [bib_ref] Multisensory integration of speech signals: the relationship between space and time, Jones [/bib_ref]. Their results showed that the facilitation effect is stronger when auditory stimuli are presented behind the participants. These results indicated that audiovisual speech perception is impervious to spatial discordances due to specialized processing of more complex speech signals [bib_ref] Audio-visual speech perception is special, Tuomainen [/bib_ref].
Moreover, [bib_ref] Cortical integration of audio-visual speech and non-speech stimuli, Wyk [/bib_ref] investigated the cortical integration of audiovisual speech and non-speech stimuli. Their findings demonstrated the differential activation of cortical networks involved in the integration of speech and non-speech stimuli [bib_ref] Cortical integration of audio-visual speech and non-speech stimuli, Wyk [/bib_ref]. However, whether simple non-speech signals originating from behind participants can enhance the detection of visual stimuli remains unclear. Second, whether simple non-speech signals originating 90u from the left or right of participants in the same horizontal plane as the ''behind'' signals can affect the detection of visual stimuli also remains unclear from electrophysiological evidence.
The present study investigates the effects of auditory stimuli in the horizontal plane on audiovisual integration using high density ERP recording and analyzes the neural bases of audiovisual integration in more detail. We designed a simple visual discrimination task that included visual stimuli, auditory stimuli and audiovisual stimuli, which were randomly presented with equal probability. Visual stimuli consisted of Gabor gratings, which were presented directly in front of the participants. Auditory stimuli consisted of a 3000 Hz sinusoidal tone, which was presented at one location on an equidistant horizontal plane at the front (0u, the fixation point), right (90u), back (180u), or left (270u) of the participants. By comparing the audiovisual integration elicited by the visual stimulus presented at the front location with the auditory stimuli presented at the four horizontal plane locations, we examined whether audiovisual integration can be modulated according to the positions of the auditory stimuli in the horizontal plane, particularly originating from behind the participants.
# Materials and methods
## Participants
Fourteen healthy volunteers (ages 21-24 years, mean age 22.5 years) participated in this study. All of the participants had normal or corrected-to-normal vision and normal hearing capabilities. The participants provided written informed consent for their participation in this study, which was previously approved by the ethics committee of Okayama University.
## Stimuli and task
The experiment was performed in a dimly lit, sound-attenuated and electrically shielded room (laboratory room, Okayama University, Japan). Streams of unimodal visual, unimodal auditory, and bimodal audiovisual stimuli (auditory and visual components that occur simultaneously) were randomly presented.
The unimodal visual stimuli consisted of two subtypes of standard and target stimuli. The visual standard stimulus was a Gabor patch with vertical gratings (2u diameter, spatial frequency = 1.6 cycles/degree). The visual target stimulus was a Gabor patch with horizontal gratings (2u diameter, spatial frequency = 1.6 cycles/degree). These visual stimuli were presented approximately 4u below the fixation point on a 21-inch computer monitor (128 background color) positioned 70 cm in the front of the subject's eyes. The visual stimulus was titrated for each subject so that detection required highly focused attention but could be performed at approximately 80% accuracy. This adjustment was accomplished by changing the contrast. The contrast of these ranged from 3% to 7% in units of Michelson contrast: (max-min)/ (max+min), with max and min being the maximal and minimal values of the Gabor patch.
The unimodal auditory stimulus was a 3000 Hz sinusoidal tone, with a linear rise and fall time of 5 milliseconds and amplitude of 60 dB. The auditory stimulus originated from one of four loudspeakers that were hidden by a black curtain. Four loudspeakers were positioned 80 cm from the front (0u, the fixation point), right (90u), back (180u), and left (270u) of the participants, at the level of the participant's ears. The bimodal audiovisual stimulus consisted of the simultaneous presentation of both the unimodal visual stimuli (standard or target) and the unimodal auditory stimuli (originating from one of the four locations). The target stimuli were presented at a frequency of approximately 19% of the total stimuli. The duration of each type of stimulus was 40 ms. The inter-stimulus interval (ISI) varied randomly between 800 ms and 1200 ms (mean ISI = 1000 ms).
Eight experimental blocks were performed in this study, with each block lasting approximately 4 min. Each block consisted of 266 trials in total (i.e., visual (24+10), auditory (24 6 4) and audiovisual (24 64+10 64)), in which the unimodal visual stimuli contained 24 standard stimuli and 10 target stimuli, unimodal auditory stimuli included 96 standard stimuli with 24 stimuli being presented at each location, and bimodal audiovisual stimuli contained 10 target stimuli and 24 standard stimuli when the auditory stimuli were presented at one of the four locations (in total, 40 target stimuli and 96 standard stimuli at four locations). These stimuli were randomly presented in each block. Each block had a 3000 ms fixation period followed by the test stimulus. During the fixation period, there was a cross (+) on the screen. During the response period that followed the test stimulus, the screen was clear. The experiment then continued with the next trial occurring at the set ISI time (800-1200 ms), regardless of whether the subject had responded to the target stimulus. Throughout the experiment, the subjects were required to fix their eyes on a centrally presented fixation point on a screen and to attend to the visual stimuli containing the unimodal visual stimulus and the visual segment of the audiovisual stimuli, while ignoring all auditory stimuli [fig_ref] Figure 1: Experimental paradigm and stimuli [/fig_ref]. The participants were instructed to respond to visual target stimuli by pressing the left button of a computer mouse with their right hand as quickly and accurately as possible.
## Apparatus
Stimulus presentation was controlled by a personal computer running Presentation software (Neurobehavioral Systems, Albany, CA). An electroencephalographic (EEG) system (BrainAmp MR plus, Gilching, Germany) was used to record EEG signals through 32 electrodes mounted on an electrode cap (Easy-cap, Herrsching Breitbrunn, Germany). All signals were referenced to the left and right earlobes. Horizontal eye movements were measured by deriving the electrooculogram (EOG) from one electrode placed at the outer canthi of the left eye. Vertical eye movements and eye blinks were detected by deriving an EOG from an electrode placed approximately one centimeter below the subject's left eye. The impedance was maintained below 5 kV. Raw signals were digitized using a sample frequency of 500 Hz with a 60 Hz notch filter, and all data were stored digitally for off-line analysis. The ERP analysis was carried out using Brain Vision Analyzer software (version 1.05, Brain Products GmbH, Munich, Bavaria, Germany).
# Data analysis
Behavioral data. Hit rates and response times to target stimuli were computed separately for each stimulus type. Hit rates were the number of correct responses to target stimuli divided by the total number of target stimuli. Response time data were analyzed for correct responses. The data for incorrect responses to standard stimuli were submitted using false alarm rates. We then used a z-transformed ratio to compute sensitivity (d') and criterion (c) [bib_ref] Calculation of signal detection theory measures, Stanislaw [/bib_ref]. Behavioral results for all measures (response times, hit rates, false alarm rates, d' and response criterion) were analyzed using a repeated-measures analysis of variance with the stimulus modalities (visual stimuli and audiovisual stimuli) as subject factors, and the alpha level was set at p,0.05.
ERP analysis. The ERPs elicited by the standard stimuli were analyzed. The EEG and EOG signals were amplified and band-pass filtered with an analog filter of 0.01-100 Hz at a sampling rate of 500 Hz. EEG and EOG signals were divided into epochs from 100 ms before the stimulus onset to 600 ms after onset, and baseline corrections were made against 2100-0 ms. Trials with vertical eye movements and eye blinks (vertical EOG amplitudes exceeding 6100 mV), horizontal eyeball movements (horizontal EOG amplitudes exceeding 625 mV), or other artifacts (a voltage exceeding 680 mV at any electrode location relative to baseline) were excluded from the analysis. These trials were subjected to automatic rejection. Responses associated with false alarms were also rejected from the analysis. The data were then averaged for each stimulus type following digital filtering using a band-pass filter of 0.01-30 Hz. The grand-averaged data were obtained across all participants for each stimulus type. Audiovisual integration was assessed by the difference wave [AV-(A+V)], obtained by subtracting the sum of the responses to the unimodal stimuli from the responses to the bimodal stimuli [bib_ref] The spatiotemporal organization of auditory, visual, and auditory-visual evoked potentials in rat..., Barth [/bib_ref] [bib_ref] Word and Nonword Repetition Withinand Across-Modality: An Event-Related Potential Study, Rugg [/bib_ref]. The logic of this additive model is that the ERPs to bimodal (AV) stimuli are equal to the sum of the unimodal (A+V) responses plus the putative neural activities specifically related to the bimodal nature of the stimuli. Mean amplitudes were calculated for all electrodes at consecutive windows of 20 ms each between stimulus onset and 500 ms after presentation of the stimulus. The mean amplitude data were analyzed using repeatedmeasures analysis of variance with the within-subjects factors of modality (AV, A+V), time-window and electrodes. The Greenhouse-Geisser Epsilon correction was applied to adjust the degrees of freedom of the F ratios as necessary. All statistical analyses were carried out using SPSS version 16.0 software package (SPSS, Tokyo, Japan).
# Results
# Behavioral results
The average reaction times are shown in [fig_ref] Table 1: Behavioral mean data over all participants in the experiment [/fig_ref]. Analysis of the factor stimulus type confirmed that reaction times to unimodal visual and bimodal audiovisual stimuli (auditory stimuli from four locations) differed significantly [F (4, 52) = 8.079; p = 0.001]. In addition, post-hoc comparisons found that the reaction times to the bimodal audiovisual stimuli in which auditory stimuli were presented in the front (p = 0.01) and back (p = 0.031) of the participants were significantly faster than those to the unimodal visual stimuli. These results indicate that a synergistic effect occurred when the auditory stimuli were presented in the back space. However, the pairwise comparisons showed that no significant differences in reaction times were found between the bimodal audiovisual stimuli in which auditory stimuli were presented on the left (p = 0.333) or right (p = 0.066) of the participants and the unimodal visual stimuli.
The hit rates showed similar effects as the response times [fig_ref] Table 1: Behavioral mean data over all participants in the experiment [/fig_ref]. These effects were statistically expressed as a main effect of the within-subject factor stimulus type [F (4, 52) = 7.317; p = 0.002]. The pairwise comparisons showed that hit rates differed significantly between unimodal visual and bimodal audiovisual stimuli in which auditory stimuli were presented at the front (p = 0.02) or back (p = 0.027) of the participants (the responses to the bimodal audiovisual stimuli were faster), but no significant differences were found between unimodal visual stimuli and bimodal audiovisual stimuli in which auditory stimuli were presented on the left (p = 0.068) or right (p = 0.336) of the participants. However, regarding the false alarm rates, there was no significant effect of the factor stimulus type [F (4, 52) = 1.328, p = 0.28]. As [fig_ref] Table 1: Behavioral mean data over all participants in the experiment [/fig_ref] shows, the false alarm rates for stimulus type were low, particularly for unimodal visual stimuli, which produced the lowest false alarm rates. The hit and false alarm rates were then used to compute the signal detection measure d'. There was no significant effect of stimulus type on d' [F (4, 52) = 0.852, p = 0.466] [fig_ref] Table 1: Behavioral mean data over all participants in the experiment [/fig_ref]. These results indicate that if the hit rate to one stimulus type was high, false alarm rate of this stimulus type was also high. Thus, no significant difference was found between stimulus types for d'. Response criterions are also shown in [fig_ref] Table 1: Behavioral mean data over all participants in the experiment [/fig_ref] , which shows that a significant effect on response criterions were found for stimulus types [F (4, 52) = 6.143, p = 0.001]. Additionally, pairwise comparisons found that response criterions to the bimodal audiovisual stimuli in which auditory stimuli were presented at the front (p = 0.019) or back (p = 0.040) of the participants were lower than those to the unimodal visual stimuli.
# Erp results
Event-related potentials: unimodal stimuli. [fig_ref] Figure 2: Waveforms and scalp topographies of unimodal stimuli [/fig_ref] shows the group-averaged ERPs to unimodal visual stimuli and unimodal auditory stimuli at a subset of selected electrode sites for which the potentials were recorded at greater amplitudes. The visual ERP showed a prominent negative wave peaking at approximately 250 ms after stimuli onset (N250 component) at occipital sites (peak: 22.63 mV at O2) [fig_ref] Figure 2: Waveforms and scalp topographies of unimodal stimuli [/fig_ref]. For the auditory stimuli, which were presented at the front (0u), back (180u), left (270u) and right (90u) of the subjects, ERPs showed a prominent negative wave, and the negative N1 wave peaked at approximately 140 ms post-stimulus at the frontal sites (peak: 22.61 mV, 22.96 mV, 24.15 mV, 24.19 mV at Fz). Significantly, scalp topographies showed that the amplitude of the auditory N1 was greater over the hemisphere contralateral in response to stimulus presentation [fig_ref] Figure 2: Waveforms and scalp topographies of unimodal stimuli [/fig_ref] , below).
# Event-related
potentials: audiovisual integration.shows the AV and (A+V) ERPs at several electrodes when auditory stimuli were presented at the front, back, left and right of the subjects. Several remarkable integration patterns were identified over the right temporal area and occipital area at approximately 160 to 200 ms when the auditory stimuli were presented in front of the subjects, and at the parietal and occipital areas at approximately 360 to 400 ms when the auditory stimuli were presented at the back of the subjects. These different integration patterns are analyzed in detail below. However, few or no significant integration patterns were found when the auditory stimuli were presented at the left or right of the subjects.
Integration in right temporal and occipital areas (160 to 200 ms). When auditory stimuli were presented in front of the subjects, the observed effects were the most notable because the differences in amplitudes between AV and (A+V) were highly significant at the right temporal area and extended to the right occipital electrodes (P4, P8, CP6, Oz, O2) ( . In addition, the topographies of the integration effect in this latency window to auditory stimuli at the back of the subjects appeared to be stronger than when the auditory stimuli originated from the other three locations [fig_ref] Figure 4: Topography of the different significant spatio-temporal patterns of integration [/fig_ref].
# Discussion
The main focus of this study was to determine whether auditory stimuli in the horizontal plane affects audiovisual integration, particularly auditory stimuli presented at the back of subjects. Our behavioral and neural results demonstrated that integration occurred when the auditory stimuli were presented at the back and front of the participants. Two main event-related potential components related to audiovisual integration were found: first, auditory stimuli from the front location produced an ERP reaction over the right temporal area and right occipital area at approximately 160-200 ms; second, auditory stimuli from the back produced a reaction over the parietal and occipital areas at approximately 360-400 ms. No significant integration was found when auditory stimuli originated from the left or right spaces.
## Integration of auditory stimuli from the front
A major integration was found in the 160-200 ms interval at the right temporal area that extended to the right occipital electrodes. Previous studies have reported audiovisual integration in which stimuli were presented in the frontal horizontal plane, and the observed reactions are closely analogous to this result [bib_ref] Early auditory-visual interactions in human cortex during nonredundant target identification, Fort [/bib_ref] [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref] [bib_ref] Effects of Spatial Congruity on Audio-Visual Multimodal Integration, Teder-Sä Lejä Rvi [/bib_ref] [bib_ref] Multisensory Interactions Elicited by Audiovisual Stimuli Presented Peripherally in a Visual Attention..., Wu [/bib_ref]. However, the latency for centrally presented stimuli at 145-160 ms occurred somewhat earlier than previously found [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref]. This effect may have been due to attention, which has been shown to augment audiovisual integration processes, is that audiovisual integration under attended condition was greater than unattended condition [bib_ref] Selective Attention and Multisensory Integration: Multiple Phases of Effects on the Evoked..., Talsma [/bib_ref] [bib_ref] Attention and the multiple stages of multisensory integration: A review of audiovisual..., Koelewijn [/bib_ref] [bib_ref] The multifaceted interplay between attention and multisensory integration, Talsma [/bib_ref] ; additionally, and audiovisual integration depends on the stimuli being fully attended [bib_ref] Selective Attention and Audiovisual Integration: Is Attending to Both Modalities a Prerequisite..., Talsma [/bib_ref]. In the current study, only centrally presented visual stimuli were attended, unlike a study by [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref] , who investigated audiovisual integration under conditions of simultaneous attention to both modalities. In addition, a clear finding from our results is the strong predominance of the right hemisphere in audiovisual integration. This result is in agreement with findings from previous studies, which showed that neural activation was greater in the right hemisphere [bib_ref] Auditory-Visual Integration during Multimodal Object Recognition in Humans: A Behavioral and Electrophysiological..., Giard [/bib_ref] [bib_ref] The physiology of coloured hearing A PET activation study of colour-word synaesthesia, Paulesu [/bib_ref] [bib_ref] Dynamics of corticosubcortical cross-modal operations involved in audio-visual object detection in humans, Fort [/bib_ref]. However, in one study in which subjects received auditory, visual, and audiovisual letters of the roman alphabet and were required to identify them regardless of stimulus modality, the left posterior superior temporal sulcus (STS) showed prominent audiovisual integration for letters [bib_ref] Audiovisual integration of letters in the human brain, Raij [/bib_ref]. Moreover, previous studies on speech stimuli have found that the left hemisphere is responsible for most integration functions [bib_ref] Spatiotemporal dynamics of audiovisual speech processing, Bernstein [/bib_ref] [bib_ref] Neural processing of asynchronous audiovisual speech perception, Stevenson [/bib_ref] [bib_ref] Perceptual Fusion and Stimulus Coincidence in the Cross-Modal Integration of Speech, Miller [/bib_ref]. These findings indicated that the left hemispheric is involved in the integration of certain types of information. Therefore, different characteristic stimulus types have significant influences on audiovisual integration between hemispheres.
## Integration of auditory stimuli from the back
The most important, novel findings of the present study were the effects of auditory stimulation originating behind the participants (180u) on visual perception. In this portion of the study, behavioral facilitation was significant (p,0.05) [fig_ref] Table 1: Behavioral mean data over all participants in the experiment [/fig_ref] , and activity in the parietal and occipital areas was identified at approximately 360-400 ms after stimulus onset with audiovisual integration [fig_ref] Figure 4: Topography of the different significant spatio-temporal patterns of integration [/fig_ref]. These results were not in agreement with several previous studies [bib_ref] The Merging of the Senses, Stein [/bib_ref] [bib_ref] Spatial and temporal factors determine auditory-visual interactions in human saccadic eye movements, Frens [/bib_ref] [bib_ref] Multisensory processing in the redundant-target effect: A behavioral and event-related potential study, Gondan [/bib_ref]. In these studies, performance enhancement of simple detection tasks was reduced or absent with spatial separation of the stimuli. In these previous studies, visual and auditory stimuli were both presented in front of the participants and were separated by only 30u or 40u. Additionally, other previous studies have investigated the perception of auditory stimuli in the horizontal and sagittal planes. Findings from these studies showed that the perception of spatial auditory stimuli can be affected by the distance between the auditory stimuli from the two ears and demonstrated that the contribution of the near ear increases and, conversely, that of the far ear decreases [bib_ref] The contribution of two ears to the perception of vertical angle in..., Morimoto [/bib_ref] [bib_ref] The influence of pinnae-based spectral cues on sound localization, Musicant [/bib_ref]. Thus, when the auditory stimuli were presented in the median vertical plane (front 0u; back 180u), the distance of the auditory signal from the front and back space to the two ears were equal, suggesting that the signal reached both ears simultaneously. Thus, the auditory event was temporally congruent with the visual signal that was presented in front of the participant. Temporal importance was also confirmed in several studies on multisensory interaction or integration, which found that the interaction or integration was greater when the visual and auditory stimuli occurred simultaneously [bib_ref] Cross-modal perceptual integration of spatially and temporally disparate auditory and visual stimuli, Lewald [/bib_ref] [bib_ref] Time-Window-of-Integration (TWIN) Model for Saccadic Reaction Time: Effect of Auditory Masker Level..., Colonius [/bib_ref]. The results also demonstrated that spatial proximity of the two stimuli was critical for integration. Thus, the delayed response in the EEG signal to auditory signal from the back space might be due to spatial discordance between the visual and auditory stimuli. Furthermore, previous functional magnetic resonance imaging (fMRI) studies also investigated the perception and detection of auditory stimuli in the horizontal directions. The behavioral data from these studies showed that no significant difference was found between the detection of auditory stimuli from the front and back spaces. However, their fMRI results showed that the activity of the left posterior temporal gyri (pSTG) was greater for back than front auditory stimuli [bib_ref] Neural correlates of sound externalization, Callan [/bib_ref]. These results indicate that integration was less efficient from the back space compared to the front space, potentially because a greater number of neural resources are needed to perform the same task. Nevertheless, when visual and auditory stimuli were simultaneously presented in front of the subjects, a spatially consistent sound might rapidly facilitate attention and perception to visual stimuli [bib_ref] Effects of Spatial Congruity on Audio-Visual Multimodal Integration, Teder-Sä Lejä Rvi [/bib_ref]. Therefore, in this study, audiovisual integration began later in the back space than in the front space according to the EEG signal, which was consistent with the behavioral results on ''back'' effects.
In a similar manner, Zampini et al. (2007) studied auditory and somatosensory stimuli pairings that were presented to either the same or different positions in the front and/or back of the participants. The results from the study showed that responses were significantly more rapid to bimodal auditory and somatosensory stimuli than to a unimodal modality, regardless of whether the stimuli were presented from the front or back spaces [bib_ref] Auditory-somatosensory multisensory interactions in front and rear space, Zampini [/bib_ref]. The findings indicated that auditory stimuli from the back could also improve the perception of the somatosensory stimuli to the same level as that achieved by auditory stimuli from the front space. These results, together with those of [bib_ref] Multisensory integration of speech signals: the relationship between space and time, Jones [/bib_ref] and the present study, confirm that the facilitation of behavioral responses occurred even though modalities were presented in the front and back locations. Thus, these results suggest that new neural brain activities might be activated when modalities are presented in the back and that integration could occur and does not completely depend on the spatial position of the stimuli. However, further electrophysiological studies are needed to confirm and elucidate these neural mechanisms. More specifically, studies on stimuli presented in the back space are often neglected in multisensory research.
## Integration of auditory stimuli from the left and right spaces
Our results showed that the contralateral auditory cortex was activated when unimodal auditory stimuli were presented in the left or right locations; this finding is consistent with previous studies [bib_ref] Lateralized auditory spatial perception and the contralaterality of cortical processing as studied..., Woldorff [/bib_ref]. However, no significant integration was found when visual stimuli were centrally presented and auditory stimuli were presented in the left (90u) or right (90u) location. Recently, behavioral studies have found that the facilitative effects of peripherally paired visual-auditory stimuli were less than those presented centrally (0u) and have even found that no facilitation was produced by the right 90u location [bib_ref] Interactions between the spatial and temporal stimulus factors that influence multisensory integration..., Stevenson [/bib_ref]. In this study, it was confirmed by ERP analysis that no significant integration occurred when auditory stimuli were presented from the lateral locations. A possible reason for this result is that the two ears, which are in lateral symmetry on the head have inter-aural differences, such as auditory stimuli originating from the right (90u) side and thus different amounts of times were needed to inputs from both the left and right ears. Thus, under this condition, the auditory signal is both temporally and spatially incongruent with the visual signal. Moreover, previous studies have found that the timing of auditory and visual stimuli has effects on integration [bib_ref] Timecourse of coactivation in bimodal divided attention, Miller [/bib_ref] [bib_ref] Multimodal access to verbal name codes, Berryhill [/bib_ref] and also confirmed that integration was greater when the auditory stimuli were presented in close temporal and spatial proximity with the visual stimuli [bib_ref] Cross-modal perceptual integration of spatially and temporally disparate auditory and visual stimuli, Lewald [/bib_ref] [bib_ref] Time-Window-of-Integration (TWIN) Model for Saccadic Reaction Time: Effect of Auditory Masker Level..., Colonius [/bib_ref]. Thus, it had been suggested that temporal and spatial processing of the two signals were pivotal in determining whether integration would occur. These findings were consistent with our results. When a visual stimulus was presented at the front and an auditory stimulus was presented at the front or back, integration or interaction occurred; however, this integration was delayed when the auditory signal was presented in the back space. However, when the auditory information was presented at the sides (and the visual signal was presented at the front), integration was not observed. Further electrophysiological studies are needed to elucidate these neural mechanisms.
# Conclusions
This study shows the influences of auditory stimuli presented in the horizontal plane at the front (0u, the fixation point), right (90u), back (180u), and left (270u) of the participant on audiovisual integration processing. The data clearly showed that audiovisual integration occurred over the right temporal area and right occipital area at approximately 160-200 ms latency when the auditory stimuli were located in front of the participant. However, no significant integration was found when auditory stimuli were presented in the right and left spaces. More specifically, when the auditory stimuli were presented from the back, audiovisual integration was observed over the parietal and occipital areas at approximately 360-400 ms. We believe that our findings, particularly regarding auditory stimuli originating from the back of participants, are likely to be a useful reference for further studies that investigate the integration mechanism when different modality stimuli are presented at different spatial positions.
[fig] Figure 1: Experimental paradigm and stimuli. (A) Subjects sat approximately 70 cm from the screen. Visual stimuli were presented on the front screen. Auditory stimuli were randomly presented at one of four locations using speakers (front 0u, right 90u, back 180u, and left 270u). (B) Stimuli types. doi:10.1371/journal.pone.0066402.g001 [/fig]
[fig] Figure 2: Waveforms and scalp topographies of unimodal stimuli. (A) Unimodal visual occipital N250 component. (B) Unimodal auditory N1 (auditory stimuli were presented at the front, back, left or right of the subjects). Note that the auditory N1 effect shows a clear contralateral enhancement (below). doi:10.1371/journal.pone.0066402.g002 Auditory Location Affects Audiovisual Integration PLOS ONE | www.plosone.org [/fig]
[fig] Figure 3: Analysis of these amplitudes showed that the main effects of the modality [F (1, 13) = 13.23, p,0.01] and electrode [F (4, 52) = 10.33, p,0.01]. However, no significant interaction between modality and electrode were found [F (4, 52) = 0.713, p = 0.532]. The difference in amplitudes between AV and (A+V) were apparent at approximately 140 ms and reached a statistical significance of 0.01 between 160 and 200 ms at P4 [F (1, 13) = 13.01, p = 0.005] (mean amplitude, AV-(A+V): 0.95 mV), P8 [F (1, 13) = 15.32, p = 0.001] (mean amplitude: 1.16 mV) and CP6 [F (1, 13) = 19.09, p = 0.003] (mean amplitude: 1.10 mV). The occipital differences between AV and (A+V) were confirmed at Oz [F (1, 13) = 10.13, p = 0.010] (mean amplitude: 0.95 mV) and O2 [F (1, 13) = 6.52, p = 0.029] (mean amplitude: 0.91 mV). However, from approximately 160 to 200 ms latency, no significant differences between AV and (A+V) were found at the right temporal and occipital areas when auditory stimuli were presented to the back [F (1, 13) = 0.09, p = 0.762], left [F (1, 13) = 2.21, p = 1.169] and right [F (1, 13) = 3.46, p = 0.093] of the subjects (Figure 3). The topography of [AV-(A+V)] showing the effects of auditory stimuli from the four different locations is shown inFigure 4. The positive values in the [AV-(A+V)] differences at the right temporal and occipital areas were due to smaller amplitudes in the AV than in the (A+V) conditions. This result indicated that the amplitude of the response in auditory N1 was smaller for bimodal stimuli than for unimodal auditory stimuli.Integration of parietal and occipital areas (360 to 400 ms). When auditory stimuli were presented behind the subjects, at approximately 360 to 400 ms latency, significant differences in the amplitudes between the AV and (A+V) conditions [F (1, 13) = 8.32, p,0.05] were found at several parietal and occipital electrodes (Pz, P4, Oz, O2) (Figure 3). Statistical significance of p,0.01 was reached between 360 and 400 ms at P4 (mean amplitude, AV-(A+V): 1.02 mV) and also between 380 and 400 ms at Oz (mean amplitude: 0.98 mV) and O2 (mean amplitude: 1.09 mV). However, neither a significant effects of electrode type (Pz, P4, Oz, O2) nor significant interactions between electrode and modality were found.In the same manner, no significant corresponding patterns were observed in this latency window when the auditory stimuli were presented at the other three locations(front [F (1, 13) = 3.41, p = 0.094], left [F (1, 13) = 0.54, p = 0.479] and right [F (1, 13) = 2.71, p = 0.131]) [/fig]
[fig] Figure 4: Topography of the different significant spatio-temporal patterns of integration. Audiovisual integration occurred (A) over the right temporal area at approximately 160-200 ms when the auditory stimuli were located in front of the participants; (B) over the right occipital area at approximately 160-200 ms when auditory stimuli were located in front of the participants; (C) over the right occipital area at approximately 360-400 ms when auditory stimuli were located in back of the participants. (A F , A B , A L , A R : unimodal auditory stimuli were presented in the front, back, left and right of the subjects, respectively). doi:10.1371/journal.pone.0066402.g004 [/fig]
[table] Table 1: Behavioral mean data over all participants in the experiment. [/table]
|
Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note
J Korean Assoc Oral Maxillofac Surg 2015;41:284-289)In cases of severe alveolar bone atrophy in the posterior maxillary area, which has only a thin sinus floor, the autogenous tooth bone graft block (ABTB) was used to wrap the implant to enhance its primary stability and osseointegration in the sinus. These cases with four years of clinical follow-up demonstrate the applicability of the ABTB in maxillary sinus membrane elevation to improve the outcomes of implant placement.
already has shown good osteoinductive, osteoconductive, and remodeling ability due to its type I collagenous nature as well as its three-dimensional scaffold [bib_ref] Microscopic feature, protein marker expression, and osteoinductivity of human demineralized dentin matrix, Park [/bib_ref] [bib_ref] Bone grafts using autogenous tooth blocks: a case series, Kim [/bib_ref]. This collagenous nature could act as an excellent vascular scaffold in the sinus cavity, which has a poor blood supply. The major role of the ABTB in this report was to wrap around the implant to enhance its primary stability and osseointegration into the sinus. This is called the "ring technique".
## Ii. technical note
This technique can be used in cases during which the residual alveolar bone height to the maxillary sinus floor is insufficient for implant stability. In our case, a 51-year-old man presented with his left maxillary first molar missing for several months. We decided to perform sinus bone graft and simultaneous implant placement. ABTB was prepared from his son's wisdom tooth because the patient did not have teeth to be extracted for manufacturing. This process met the standards of the technical manual of the Korea Tooth Bank, as has been previously reported [bib_ref] Tooth bank system for bone regeneration: safety report, Kim [/bib_ref] [bib_ref] Familial tooth bone graft: case reports, Lee [/bib_ref]. The extracted wisdom tooth was immersed in 75% alcohol and then the soft tissues and calculus were removed. Crowns were severed at the cement enamel junctions and ABTBs were made from the root
# I. introduction
There have been considerable efforts in basic and clinical research to find the best bone graft material for implant osseointegration in the sinus. The autogenous tooth bone graft (AutoBT; Korea Tooth Bank, Seoul, Korea) was firstly developed in 2008. There have been several clinical studies evaluating the use of AutoBT in the sinus, with all results showing more favorable outcomes in bone regeneration as compared to other conventional graft materials 1,2 . The autogenous tooth bone graft block (ABTB), which is made from root dentin, has been shown to be both clinically safe and efficacious in several studies. In particular, the ABTB has been reported to be an appropriate material for socket preservation, guided bone regeneration (GBR), and ridge augmentation. ABTB to extraorally fit into the ABTB through the pulp chamber as tightly as possible. With this procedure, we could secure additional stability of the thin alveolar bone. After inserting the ABTB upside down into the sinus, the implant was placed through the thin sinus floor and into the apex of the ABTB while being grabbed by a small Kelly or mosquito clamp to provide stability without additional fixation. We could see blood pooling of the ABTB, which met the wet-portion of the tooth. Additional holes sized 0.2 mm were made at the surface of the canal area to create macropores for promoting vascular invasion and bone formation. Through the routine lateral window approach, we lifted the sinus membrane to identify the thin sinus floor, locate the implant hole, and spare the space for the ABTB. Before placing the ABTB into the sinus, the implant was adapted
## Kwang-ho lee et al: autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. j korean assoc oral maxillofac surg 2015
In another case, a 27-year-old man presented with a missing #16 tooth for several years without any restoration, so that only a thin sinus floor remained due to severe pneumatization. The #16 area became an anatomic defect of the alveolar bone with about 1 to 2 mm of residual bone height. In order to install the implant in a single stage, sinus augmentation was inevitable and an ABTB was chosen to obtain initial stability of the implant on the thin alveolar bone. An tability requirements of bone graft substitutes. [fig_ref] Figure 5: The autogenous tooth bone graft block was adapted into the sinus [/fig_ref] The lateral wall was then replaced in situ. [fig_ref] Figure 7: The lateral window was repositioned [/fig_ref] When compared to the X-rays obtained immediately after the operation, the X-rays after four years showed well-formed bone surrounding the implant. The border between the implant and ABTB almost disappeared, which means the ABTB was successfully incorporated. Moreover, it was difficult to identify any marginal bone loss even with the final prosthesis. [fig_ref] Figure 8: The cone-beam computed tomography immediately after the simultaneous installation of the implant... [/fig_ref]
## Kwang-ho lee et al: autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. j korean assoc oral maxillofac surg 2015
## Kwang-ho lee et al: autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. j korean assoc oral maxillofac surg 2015
## Kwang-ho lee et al: autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. j korean assoc oral maxillofac surg 2015
## Kwang-ho lee et al: autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. j korean assoc oral maxillofac surg 2015
tion. [fig_ref] Figure 6: The implant was simultaneously installed through the alveolar bone and into the... [/fig_ref] III. Discussion reported that autogenous demineralized dentin matrix could be used for sinus augmentation, there have been several reports of the use of AutoBT in the sinuses. Lee et al. [bib_ref] Porcine study on the efficacy of autogenous tooth bone in the maxillary..., Lee [/bib_ref] compared the use of AutoBT powder with synthetic ABTB was prepared using his lower right impacted mandibular third molar. A pulp chamber that is too small to accommodate an implant can be enlarged with an instrument to fit the implant. The sinus membrane was elevated and simultaneous implant placement was performed using the ABTB graft similar to the case above. [fig_ref] Figure 5: The autogenous tooth bone graft block was adapted into the sinus [/fig_ref] Four years later, we could see that the implant osseointegrated with the final prosthesis. The whole bone housing was intact and there was no marginal bone resorption around the implant neck area when compared to the films immediately after the opera-
## Kwang-ho lee et al: autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. j korean assoc oral maxillofac surg 2015
HA in the sinus in both prospective and retrospective human clinical trials, the authors reported that AutoBT could be an alternative to synthetic bone grafts in sinus lifts based on the observation that the average bone resorption with AutoBT powder was similar to that of osteon composites 1 year after surgery with no serious complications or implant failures 1,2 . Kim et al. [bib_ref] Bone grafts using autogenous tooth blocks: a case series, Kim [/bib_ref] reported the clinical results of 12 cases involving ABTB, which included GBR, ridge augmentation, extraction socket graft, and sinus bone graft. Among these cases, only 3 were in the sinus, though they showed that ABTB could be hydroxyapatite (HA) in the sinus of miniature pigs. Histomorphometrical analysis showed that more new bone was formed in the AutoBT group than the synthetic HA group. In a clinical trial, Kim et al. [bib_ref] Evaluation of the healing process of autogenous tooth bone graft material nine..., Kim [/bib_ref] reported that the AutoBT particles united well with the new bone in the sinus, with bone formation starting from the alveolus and gradually rising toward the sinus. When comparing the use of AutoBT with synthetic
[fig] Figure 1 2015, Figure 2 2015, Figure 3 2015, Figure 4: The cone-beam computed tomography of the sinus before the operation. The residual bone height was about 1 to 2 mm on #26.Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac SurgThe occlusal view of the autogenous tooth bone graft block, which was made from the root portion of the extracted tooth.Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac SurgThe sinus lateral window was opened and the membrane was elevated.Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac SurgThe autogenous tooth bone graft block was adapted onto the implant. [/fig]
[fig] Figure 5: The autogenous tooth bone graft block was adapted into the sinus. [/fig]
[fig] Figure 6: The implant was simultaneously installed through the alveolar bone and into the autogenous tooth bone graft block. [/fig]
[fig] Figure 7: The lateral window was repositioned. [/fig]
[fig] Figure 8: The cone-beam computed tomography immediately after the simultaneous installation of the implant and the autogenous tooth bone graft block. [/fig]
[fig] Figure 9 2015, Figure 12: The cone-beam computed tomography after four years. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg 2015 Fig. 10. The preoperative X-ray in a 27-year-old male patient with the loss of #16. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg The autogenous tooth bone graft block was used to adjust the size of the implant. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg 2015 Fig. 11. The occlusal view of the autogenous tooth bone graft block. [/fig]
[fig] Figure 13: The lateral sinus window opening and the preparation of the implant site on the alveolar bone. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg 2015 Fig. 14. The insertion of the autogenous tooth bone graft block (ABTB) into the sinus and the installation of the implant through the thin alveolar bone to the ABTB. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg 2015 Fig. 15. The positioned lateral window. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg 2015 Fig. 16. Cone-beam computed tomography immediately after implant installation with the autogenous tooth bone graft block. Kwang-Ho Lee et al: Autogenous tooth bone graft block for sinus augmentation with simultaneous implant installation: a technical note. J Korean Assoc Oral Maxillofac Surg 2015 [/fig]
|
Stability of Ion Exchange Membranes in Electrodialysis
During electrodialysis the ion exchange membranes are affected by such factors as passage of electric current, heating, tangential flow of solution and exposure to chemical agents. It can potentially cause the degradation of ion exchange groups and of polymeric backbone, worsening the performance of the process and necessitating the replacement of the membranes. This article aims to review how the composition and the structure of ion exchange membranes change during the electrodialysis or the studies imitating it.
# Introduction
Since the 1950s, electrodialysis has been used industriallyfor the purification, separation or concentration of substances. This method is based on the transport of ions through selectively permeable membranes under the action of an applied electric field [bib_ref] Electrodialysis, a mature technology with a multitude of new applications, Strathmann [/bib_ref]. Typically, ion exchange membranes, the materials that contain at least one ion exchange polymer bearing polar groups, are used in electrodialysis. In addition to the ion exchange polymer, ion exchange membranes may contain other components; additives that increase the mechanical strength of membranes [bib_ref] Ion-exchange membrane reinforcing, Křivčík [/bib_ref] are most often used; however, other substances are also introduced, for example, nanotubes [bib_ref] Advancing the conductivity-permselectivity tradeoff of electrodialysis ion-exchange membranes with sulfonated CNT nanocomposites, Fan [/bib_ref] or oxide nanoparticles that increase membrane selectivity [bib_ref] Physicochemical and electrochemical characterization of Nafion-type membranes with embedded silica nanoparticles: Effect..., Porozhnyy [/bib_ref] [bib_ref] Influence of titanium dioxide particles percentage in modifying layer on surface properties..., Gil [/bib_ref]. The main types of ion exchange membranes are monopolar ones that have selectivity towards ions of a certain sign of charge, which are further subdivided into cation exchange membranes and anion exchange membranes, and bipolar ones, in which cation exchange and anion exchange components are simultaneously present and which are therefore impermeable to salt ions but are designed to generate H + and OH − ions at the boundary between the cation exchange and anion exchange components, which is called the bipolar boundary [bib_ref] Electric field effects on proton transfer between ionizable groups and water in..., Simons [/bib_ref] [bib_ref] Dissociation of water molecules in systems with ion-exchange membranes, Zabolotskii [/bib_ref] [bib_ref] Bipolar membranes: A review on principles, latest developments, and applications, Pärnamäe [/bib_ref]. There are also asymmetric bipolar membranes that combine the function of generation of H + and OH − ions with the function of transporting salt ions [bib_ref] Asymmetric bipolar membranes in acid−base electrodialysis, Wilhelm [/bib_ref] [bib_ref] Asymmetric bipolar membrane: A tool to improve product purity, Balster [/bib_ref] [bib_ref] Water splitting and transport of ions in electromembrane system with bilayer ion-exchange..., Melnikov [/bib_ref] [bib_ref] Antipolarization Membrane Having Anionic and Cationic Areas, Kollsman [/bib_ref] [bib_ref] Cationic-Anionic Ion-Exchange Membrane, Leitz [/bib_ref]. Scheme of electrodialysis apparatus for desalination/concentration or for production of acid and alkali is depicted in [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref]. The cost of ion exchange membranes can be a significant part of the installation costs and, depending on the salinity of the treated water, a significant part of the total cost of the produced treated water [bib_ref] Assessment of electrodialysis water desalination process costs, Strathmann [/bib_ref]. In this regard, there is interest both in reducing the cost of membranes and in extending their service life. Unfortunately, during electrodialysis the membranes are affected by a number of factors that can lead to structural damage and degradation of membrane properties including the precipitation of organic and inorganic substances [bib_ref] Concentration effects of calcium ion on polyacrylamide fouling of ion-exchange membrane in..., Zhang [/bib_ref] , shift of pH [bib_ref] Concentration polarization and water dissociation in ion-exchange membrane electrodialysis. Mechanism of water..., Tanaka [/bib_ref] [bib_ref] Effect of Alkali Treatment on the Mechanical Properties of Anion-Exchange Membranes with..., Doi [/bib_ref] [bib_ref] Alkaline stability of poly(phenylene)-based anion exchange membranes with various cations, Hibbs [/bib_ref] and interaction with the components of the treated solution [bib_ref] Sorption of pesticide endosulfan by electrodialysis membranes, Banasiak [/bib_ref] [bib_ref] Treatment of molybdate solutions by electrodialysis: The effect of pH and current..., Marder [/bib_ref]. This review is devoted to the stability of ion exchange membranes during electrodialysis and considers the factors that act during electrodialysis and affect the properties of the membrane.
## Fouling
## Definition and classification
Fouling [bib_ref] A review on ion-exchange membrane fouling during the electrodialysis process in the..., Dammak [/bib_ref] is the chemical interaction of the membrane with the components of the treated solution that leads to the deposition of some compounds, causing deterioration of the membrane properties. Usually, the deterioration of the properties of an ion exchange membrane is expressed as the loss of electrical conductivity [bib_ref] Structural and physicochemical investigation of ageing of ion-exchange membranes in electrodialysis for..., Ghalloussi [/bib_ref] or in a decrease in the limiting current density; a model description of the decrease in the limiting current depending on the type of formed fouling film is presented in. Fouling is recognized as one of the main problems of membrane technologies [bib_ref] In-depth on the fouling and antifouling of ion-exchange membranes, Dammak [/bib_ref] , and more publications are devoted to this problem than to other mechanisms of degradation of membrane properties during electrodialysis.
There are several variants for classifying the types of fouling depending on the causing agents, for example, Mikhaylin and Bazinet [bib_ref] Fouling on ion-exchange membranes: Classification, characterization and strategies of prevention and control, Mikhaylin [/bib_ref] distinguish [fig_ref] Figure 2: Types of fouling [/fig_ref] fouling with inorganic substances (which is denoted as scaling in some other articles), fouling with organic molecules and colloids (colloidal fouling of anion exchange membranes is discussed in a large number of publications, for example, in [bib_ref] Analysis of fouling potential in the electrodialysis process in the presence of..., Lee [/bib_ref] [bib_ref] Characterization of bulk and surface properties of anion-exchange membranes in initial stages..., Sarapulova [/bib_ref] [bib_ref] Unravelling the fouling behavior of anion-exchange membrane (AEM) by organic solute of..., Xiang [/bib_ref] ; an example of the formation of peptide layers on a cation exchange membrane is described in [bib_ref] Formation of peptide layers and adsorption mechanisms on a negatively charged cation-exchange..., Persico [/bib_ref] ; peptide fouling of bipolar membranes is described in [bib_ref] Impact of conductivity on the performances of electro-acidification and enzymatic hydrolysis phases..., Abou-Diab [/bib_ref] and biofouling, which, however, is more actively discussed in the context of reverse electrodialysis [bib_ref] Biofouling phenomena on anion exchange membranes under the reverse electrodialysis process, Vaselbehagh [/bib_ref] [bib_ref] A study on biofouling and cleaning of anion exchange membranes for reverse..., Tiago [/bib_ref]. The cost of ion exchange membranes can be a significant part of the installation costs and, depending on the salinity of the treated water, a significant part of the total cost of the produced treated water [bib_ref] Assessment of electrodialysis water desalination process costs, Strathmann [/bib_ref]. In this regard, there is interest both in reducing the cost of membranes and in extending their service life. Unfortunately, during electrodialysis the membranes are affected by a number of factors that can lead to structural damage and degradation of membrane properties including the precipitation of organic and inorganic substances [bib_ref] Concentration effects of calcium ion on polyacrylamide fouling of ion-exchange membrane in..., Zhang [/bib_ref] , shift of pH [bib_ref] Concentration polarization and water dissociation in ion-exchange membrane electrodialysis. Mechanism of water..., Tanaka [/bib_ref] [bib_ref] Effect of Alkali Treatment on the Mechanical Properties of Anion-Exchange Membranes with..., Doi [/bib_ref] [bib_ref] Alkaline stability of poly(phenylene)-based anion exchange membranes with various cations, Hibbs [/bib_ref] and interaction with the components of the treated solution [bib_ref] Sorption of pesticide endosulfan by electrodialysis membranes, Banasiak [/bib_ref] [bib_ref] Treatment of molybdate solutions by electrodialysis: The effect of pH and current..., Marder [/bib_ref]. This review is devoted to the stability of ion exchange membranes during electrodialysis and considers the factors that act during electrodialysis and affect the properties of the membrane.
## Fouling
## Definition and classification
Fouling [bib_ref] A review on ion-exchange membrane fouling during the electrodialysis process in the..., Dammak [/bib_ref] is the chemical interaction of the membrane with the components of the treated solution that leads to the deposition of some compounds, causing deterioration of the membrane properties. Usually, the deterioration of the properties of an ion exchange membrane is expressed as the loss of electrical conductivity [bib_ref] Structural and physicochemical investigation of ageing of ion-exchange membranes in electrodialysis for..., Ghalloussi [/bib_ref] or in a decrease in the limiting current density; a model description of the decrease in the limiting current depending on the type of formed fouling film is presented in. Fouling is recognized as one of the main problems of membrane technologies [bib_ref] In-depth on the fouling and antifouling of ion-exchange membranes, Dammak [/bib_ref] , and more publications are devoted to this problem than to other mechanisms of degradation of membrane properties during electrodialysis.
There are several variants for classifying the types of fouling depending on the causing agents, for example, Mikhaylin and Bazinet [bib_ref] Fouling on ion-exchange membranes: Classification, characterization and strategies of prevention and control, Mikhaylin [/bib_ref] distinguish [fig_ref] Figure 2: Types of fouling [/fig_ref] fouling with inorganic substances (which is denoted as scaling in some other articles), fouling with organic molecules and colloids (colloidal fouling of anion exchange membranes is discussed in a large number of publications, for example, in [bib_ref] Analysis of fouling potential in the electrodialysis process in the presence of..., Lee [/bib_ref] [bib_ref] Characterization of bulk and surface properties of anion-exchange membranes in initial stages..., Sarapulova [/bib_ref] [bib_ref] Unravelling the fouling behavior of anion-exchange membrane (AEM) by organic solute of..., Xiang [/bib_ref] ; an example of the formation of peptide layers on a cation exchange membrane is described in [bib_ref] Formation of peptide layers and adsorption mechanisms on a negatively charged cation-exchange..., Persico [/bib_ref] ; peptide fouling of bipolar membranes is described in [bib_ref] Impact of conductivity on the performances of electro-acidification and enzymatic hydrolysis phases..., Abou-Diab [/bib_ref] and biofouling, which, however, is more actively discussed in the context of reverse electrodialysis [bib_ref] Biofouling phenomena on anion exchange membranes under the reverse electrodialysis process, Vaselbehagh [/bib_ref] [bib_ref] A study on biofouling and cleaning of anion exchange membranes for reverse..., Tiago [/bib_ref]. [fig_ref] Figure 2: Types of fouling [/fig_ref]. Types of fouling. Reprinted with permission from [bib_ref] Fouling on ion-exchange membranes: Classification, characterization and strategies of prevention and control, Mikhaylin [/bib_ref]. Copyright © 2015 Elsevier B.V.
## Scaling
Fouling with inorganic substances is also referred as scaling in some sources. Its characteristic manifestation is the deposition of inorganic substances on the surface or inside the pores of membranes with the formation of a "scaly" precipitate. When scaling occurs inside the pores of the membrane, the pores can be partially or completely blocked, which reduces the electrical conductivity; in addition, with the further course of the process, the pores can be stretched by sediment, which worsens the mechanical properties of the membrane. Precipitates are usually formed by CaCO3, CaSO4and other insoluble sulfates (BaSO4, more rarely SrSO4 and RaSO4) [bib_ref] Sulfate mineral scaling: From fundamental mechanisms to control strategies, Cao [/bib_ref] , Ca(OH)2 and Mg(OH)2 [bib_ref] Mitigation of membrane scaling in electrodialysis by electroconvection enhancement, pH adjustment and..., Andreeva [/bib_ref] , which are sparingly soluble compounds; therefore, it is noteworthy that that the formation of these precipitates can occur not only in the concentration chamber, but also in the desalination chamber. An example of the precipitates formed at the side of MA-41P membrane [bib_ref] Efficient anion-exchange membranes with anti-scaling properties obtained by surface modification of commercial..., Butylskii [/bib_ref] in a solution of the following composition (Li +~0 .25 g/L, Na +~1 .15 g/L, K +~2 .5 g/L, Ca 2+~1 .5 g/L, Mg 2+~0 .3 g/L, HCO3 −~4 .3 × 10 −3 g/L, Cl −~8 .86 g/L and pH 3.9) facing the desalination chamber is shown in .
## Scaling
Fouling with inorganic substances is also referred as scaling in some sources. Its characteristic manifestation is the deposition of inorganic substances on the surface or inside the pores of membranes with the formation of a "scaly" precipitate. When scaling occurs inside the pores of the membrane, the pores can be partially or completely blocked, which reduces the electrical conductivity; in addition, with the further course of the process, the pores can be stretched by sediment, which worsens the mechanical properties of the membrane. Precipitates are usually formed by CaCO 3, CaSO 4and other insoluble sulfates (BaSO 4 , more rarely SrSO 4 and RaSO 4 ) [bib_ref] Sulfate mineral scaling: From fundamental mechanisms to control strategies, Cao [/bib_ref] , Ca(OH) [bib_ref] Electrodialysis, a mature technology with a multitude of new applications, Strathmann [/bib_ref] and Mg(OH) 2 [bib_ref] Mitigation of membrane scaling in electrodialysis by electroconvection enhancement, pH adjustment and..., Andreeva [/bib_ref] , which are sparingly soluble compounds; therefore, it is noteworthy that that the formation of these precipitates can occur not only in the concentration chamber, but also in the desalination chamber. An example of the precipitates formed at the side of MA-41P membrane [bib_ref] Efficient anion-exchange membranes with anti-scaling properties obtained by surface modification of commercial..., Butylskii [/bib_ref] in a solution of the following composition (Li +~0 .25 g/L, Na +~1 .15 g/L, K +~2 .5 g/L, Ca 2+~1 .5 g/L, Mg 2+~0 .3 g/L, HCO 3 −~4 .3 × 10 −3 g/L, Cl −~8 .86 g/L and pH 3.9) facing the desalination chamber is shown in . Reprinted with permission from [bib_ref] Fouling on ion-exchange membranes: Classification, characterization and strategies of prevention and control, Mikhaylin [/bib_ref]. Copyright © 2015 Elsevier B.V.
## Scaling
Fouling with inorganic substances is also referred as scaling in some sources. Its characteristic manifestation is the deposition of inorganic substances on the surface or inside the pores of membranes with the formation of a "scaly" precipitate. When scaling occurs inside the pores of the membrane, the pores can be partially or completely blocked, which reduces the electrical conductivity; in addition, with the further course of the process, the pores can be stretched by sediment, which worsens the mechanical properties of the membrane. Precipitates are usually formed by CaCO3, CaSO4and other insoluble sulfates (BaSO4, more rarely SrSO4 and RaSO4) [bib_ref] Sulfate mineral scaling: From fundamental mechanisms to control strategies, Cao [/bib_ref] , Ca(OH)2 and Mg(OH)2 [bib_ref] Mitigation of membrane scaling in electrodialysis by electroconvection enhancement, pH adjustment and..., Andreeva [/bib_ref] , which are sparingly soluble compounds; therefore, it is noteworthy that that the formation of these precipitates can occur not only in the concentration chamber, but also in the desalination chamber. An example of the precipitates formed at the side of MA-41P membrane [bib_ref] Efficient anion-exchange membranes with anti-scaling properties obtained by surface modification of commercial..., Butylskii [/bib_ref] in a solution of the following composition (Li +~0 .25 g/L, Na +~1 .15 g/L, K +~2 .5 g/L, Ca 2+~1 .5 g/L, Mg 2+~0 .3 g/L, HCO3 −~4 .3 × 10 −3 g/L, Cl −~8 .86 g/L and pH 3.9) facing the desalination chamber is shown in . Fouling intensity depends on the compatibility of the chemical nature of membranes and contaminants: for example, since large molecules in the food industry are usually negatively charged, in treatment of solutions of food industry anion exchange membranes are more susceptible to fouling than cation exchange ones (as, for example, shown for the processing of food industry solutions containing polyphenols [bib_ref] Characterization and cleaning of anion-exchange membranes used in electrodialysis of polyphenol-containing food..., Bdiri [/bib_ref] and liquid digestate [bib_ref] Desalination, nutrients recovery, or products extraction: Is electrodialysis an effective way to..., Wang [/bib_ref] ; since large organic molecules in the food industry are mainly hydrophobic, hydrophobic membranes are more prone to fouling than hydrophilic ones [bib_ref] Modification of an anion exchange membrane based on rapid mussel-inspired deposition for..., Zhao [/bib_ref] ; if the cause of fouling is a compound with benzene rings, such as aromatic amino acid, phenol, anthocyanin or tannin, then polystyrene-divinylbenzene membranes are more susceptible to fouling due to the occurrence of π-stacking interactions between foulants and membranes [bib_ref] Fouling of anion-exchange membranes in electrodialysis of aromatic amino acid solution, Bukhovets [/bib_ref] [bib_ref] Desalination of neutral amino acid solutions in an electromembrane system, Eliseeva [/bib_ref] than the aliphatic membranes. In addition, even the result of colloidal fouling of paired cation exchange and anion exchange membranes in the same chamber of the electrodialyzer depends on the nature of the membrane: at the end of period of useful operation in the food industry both cation exchange and anion exchange membranes were degraded and both of them lost part of their exchange capacity, but cation exchange membranes became denser and their thickness decreased while the thickness of anion exchange membranes doubled [bib_ref] Ageing of ion-exchange membranes in electrodialysis: A structural and physicochemical investigation, Ghalloussi [/bib_ref]. Fouling intensity depends on the compatibility of the chemical nature of membranes and contaminants: for example, since large molecules in the food industry are usually negatively charged, in treatment of solutions of food industry anion exchange membranes are more susceptible to fouling than cation exchange ones (as, for example, shown for the processing of food industry solutions containing polyphenols [bib_ref] Characterization and cleaning of anion-exchange membranes used in electrodialysis of polyphenol-containing food..., Bdiri [/bib_ref] and liquid digestate [bib_ref] Desalination, nutrients recovery, or products extraction: Is electrodialysis an effective way to..., Wang [/bib_ref] ; since large organic molecules in the food industry are mainly hydrophobic, hydrophobic membranes are more prone to fouling than hydrophilic ones [bib_ref] Modification of an anion exchange membrane based on rapid mussel-inspired deposition for..., Zhao [/bib_ref] ; if the cause of fouling is a compound with benzene rings, such as aromatic amino acid, phenol, anthocyanin or tannin, then polystyrene-divinylbenzene membranes are more susceptible to fouling due to the occurrence of π-stacking interactions between foulants and membranes [bib_ref] Fouling of anion-exchange membranes in electrodialysis of aromatic amino acid solution, Bukhovets [/bib_ref] [bib_ref] Desalination of neutral amino acid solutions in an electromembrane system, Eliseeva [/bib_ref] than the aliphatic membranes. In addition, even the result of colloidal fouling of paired cation exchange and anion exchange membranes in the same chamber of the electrodialyzer depends on the nature of the membrane: at the end of period of useful operation in the food industry both cation exchange and anion exchange membranes were degraded and both of them lost part of their exchange capacity, but cation exchange membranes became denser and their thickness decreased while the thickness of anion exchange membranes doubled [bib_ref] Ageing of ion-exchange membranes in electrodialysis: A structural and physicochemical investigation, Ghalloussi [/bib_ref].
In the case of colloidal fouling the formation of a bipolar boundary between the film and the membrane formed from the side of the desalination chamber also becomes a problem, which, aside from the typical consequences of fouling, causes an increase in the generation of H + and OH − ions and the issues related to this process, for example, an intensification of the precipitation of colloids of weak electrolytes. Fouling of an anion exchange membrane and anion exchange resin with single-stranded DNA as an example of a large molecule carrying a negative charge was studied in [bib_ref] Fouling of a heterogeneous anion-exchange membrane and single anionexchange resin particle by..., Belloň [/bib_ref] , and it was shown that Membranes 2023, 13, 52 5 of 18 the fouling significantly changed the current-voltage curves of the membrane and that in the case of ion exchange resin the DNA changed the electrokinetic picture completely and very likely created bipolar junction capable of water splitting.
# Fouling-conclusions
An analysis of publications on the fouling of ion exchange membranes during electrodialysis shows that the topic of organic fouling, in particular colloidal fouling of anion exchange membranes, currently presents the greatest interest, which may be due to the increasing use of electrodialysis in the food industry and, consequently, due to chemistry of solutions in the food industry where large negatively charged molecules such as pectins. A much smaller number of publications is devoted separately to scaling, and a biofouling seems to fall within the scope of related processes such as reverse electrodialysis. At the same time, evidence appeared that a simultaneous presence of large organic molecules, especially natural organic matter, and polyvalent ions aggravates fouling, as it was shown for humic acid and Ca 2+ . This problem affects not only electrodialysis but also other membrane methods that deal with surface waters and groundwater and the solution of the problem is important for feasibility of salinity gradient power generation, so it can be expected that this topic would gain wider attention in future studies.
## Other mechanisms of membrane degradation
There are also mechanisms of degradation of ion exchange membranes that are not related to the deposition of substances on the surface or in the bulk of the membranes. The effect of physicochemical properties of treated solution such as high ionic strength and chemical reactions of components of the membrane with components of solution or with H + and OH − ions that may both be initially present in the solution and be generated during the electrodialysis in a reaction on the membrane surface catalyzed by its polar groups [bib_ref] Dissociation of water molecules in systems with ion-exchange membranes, Zabolotskii [/bib_ref] cannot be reduced just to formation of sediments but has a significant effect.
## Chemical reactions
The most notable chemical reaction is concentration of H + and OH − ions in the treated solutions can be increased both due to the initially existing acidity or alkalinity of the treated solution. The problem is inherent for alkaline fuel cells that contain anion exchange membranes, where it received wide discussion [bib_ref] Density functional theory study on the degradation of fuel cell anion exchange..., Espiritu [/bib_ref] , In addition, the generation of H + and OH − ions can be enhanced by the presence of ampholytes in the solution (in [bib_ref] Generation of H+ and OH− ions in anion-exchange membrane/ampholyte-containing solution systems: A..., Pismenskaya [/bib_ref] such enhancement was studied for solutions of acid salts of weak acids).
The influence of H + and OH − ions is multifaceted. The pH shift caused by the presence of these ions leads to the loss of charge of weakly basic groups in strongly alkaline media (deprotonation of anion exchange membrane [bib_ref] Modelling of anion-exchange membrane transport properties with taking into account the change..., Kozmai [/bib_ref] and by weakly acidic groups in strongly acidic media. The deprotonation of anion exchange membranes is much more often discussed than the discharge of cation exchange membranes, which, apparently, is associated with the greater prevalence of strongly dissociating sulfonic acid membranes compared to weakly acidic membranes, with the use of weakly basic membranes containing amino groups as well as the presence of amino groups even in strongly basic membranes [bib_ref] Monovalent-ion-selective membranes for reverse electrodialysis, Güler [/bib_ref]. The discharge of functional groups of monopolar membranes leads to loss of the membrane selectivity.
In addition to reversible reactions of protonation and deprotonation, H + and OH − ions are able to enter into irreversible reactions with the membrane material. The presence of OH − ions is a much bigger issue. These ions are capable of reacting with ammonium bases and amino groups, as well as with sulfonium and phosphonium groups [bib_ref] A review on recent developments of anion exchange membranes for fuel cells..., Maurya [/bib_ref] , which eliminate a carbon chain from the heteroatom (by mechanisms of Hofmann elimination [bib_ref] Degradation of radiation grafted hydroxide anion exchange membrane immersed in neutral pH:..., Espiritu [/bib_ref] , Stevens elimination and by other mechanisms [bib_ref] Mechanism of tetraalkylammonium headgroup degradation in alkaline fuel cell membranes, Chempath [/bib_ref] , examples are given if [fig_ref] Figure 5: Scheme of the mechanisms of degradation of ammonium bases [/fig_ref].
In addition to reversible reactions of protonation and deprotonation, H + and OHions are able to enter into irreversible reactions with the membrane material. The presence of OHions is a much bigger issue. These ions are capable of reacting with ammonium bases and amino groups, as well as with sulfonium and phosphonium groups [bib_ref] A review on recent developments of anion exchange membranes for fuel cells..., Maurya [/bib_ref] , which eliminate a carbon chain from the heteroatom (by mechanisms of Hofmann elimination [bib_ref] Degradation of radiation grafted hydroxide anion exchange membrane immersed in neutral pH:..., Espiritu [/bib_ref] , Stevens elimination and by other mechanisms [bib_ref] Mechanism of tetraalkylammonium headgroup degradation in alkaline fuel cell membranes, Chempath [/bib_ref] , examples are given if [fig_ref] Figure 5: Scheme of the mechanisms of degradation of ammonium bases [/fig_ref]. In terminal cases of these reactions the polar group is completely cleaved off [bib_ref] Change of anion exchange membranes in an aqueous sodium hydroxide solution at..., Sata [/bib_ref] , causing a loss of exchange capacity and an increase in membrane hydrophobicity, which in turn can enhance membrane fouling by hydrophobic molecules [bib_ref] Alkali attack on cation-exchange membranes with polyvinyl chloride backing and binder: Comparison..., Doi [/bib_ref]. Such a reaction requires a high alkali concentration and heating (in the article [bib_ref] Alkali attack on cation-exchange membranes with polyvinyl chloride backing and binder: Comparison..., Doi [/bib_ref] , the degradation of groups containing quaternary nitrogen was studied in 3-9 M NaOH solution at 75 - C). In incomplete cases of transformation of nitrogen compounds quaternary ammonium bases are transformed into various amino groups [bib_ref] Electroconvection in systems with heterogeneous ion-exchange membranes after thermal modification, Vasil'eva [/bib_ref]. Since amino groups are more catalytically active in the reaction of generation of H + and OH − ions [bib_ref] Dissociation of water molecules in systems with ion-exchange membranes, Zabolotskii [/bib_ref] , a positive feedback loop arises that accelerates the degradation of ammonium bases. Vasil'eva et al. studied the properties of a strongly basic membrane and showed that its quaternary ammonium bases are transformed into tertiary amino groups during electrodialysis [bib_ref] Electroconvection in systems with heterogeneous ion-exchange membranes after thermal modification, Vasil'eva [/bib_ref].
Hydroxide ions can also damage the polymer matrix of the membrane. It was shown in [bib_ref] Alkali attack on cation-exchange membranes with polyvinyl chloride backing and binder: Comparison..., Doi [/bib_ref] [bib_ref] Alkali attack on anion exchange membranes with PVC backing and binder: II..., Doi [/bib_ref] that OH − ions reacts with polyvinyl chloride, which in the studied cases was represented by a reinforcing component of Neosepta AMX anion exchange membrane [fig_ref] Figure 6: A color change trend of commercial AMX during the alkali immersion test [/fig_ref] and Neosepta CMX cation exchange membrane. The reaction splat off hydrogen chloride from the polymer and generated polyenes. These compounds lower the Young's modulus of the membrane and increase its water content. In the aforementioned works it was shown that PVC of the Neosepta AMX membrane enters the cleavage reaction in much milder conditions compared to the PVC of the Neosepta CMX membrane, which is associated with the Donnan exclusion of hydroxide ions from the matrix of the cation exchange membrane.
In terminal cases of these reactions the polar group is completely cleaved off [bib_ref] Change of anion exchange membranes in an aqueous sodium hydroxide solution at..., Sata [/bib_ref] , causing a loss of exchange capacity and an increase in membrane hydrophobicity, which in turn can enhance membrane fouling by hydrophobic molecules [bib_ref] Alkali attack on cation-exchange membranes with polyvinyl chloride backing and binder: Comparison..., Doi [/bib_ref]. Such a reaction requires a high alkali concentration and heating (in the article [bib_ref] Alkali attack on cation-exchange membranes with polyvinyl chloride backing and binder: Comparison..., Doi [/bib_ref] , the degradation of groups containing quaternary nitrogen was studied in 3-9 M NaOH solution at 75 °C). In incomplete cases of transformation of nitrogen compounds quaternary ammonium bases are transformed into various amino groups [bib_ref] Electroconvection in systems with heterogeneous ion-exchange membranes after thermal modification, Vasil'eva [/bib_ref]. Since amino groups are more catalytically active in the reaction of generation of H + and OHions [bib_ref] Dissociation of water molecules in systems with ion-exchange membranes, Zabolotskii [/bib_ref] , a positive feedback loop arises that accelerates the degradation of ammonium bases. Vasil'eva et al. studied the properties of a strongly basic membrane and showed that its quaternary ammonium bases are transformed into tertiary amino groups during electrodialysis [bib_ref] Electroconvection in systems with heterogeneous ion-exchange membranes after thermal modification, Vasil'eva [/bib_ref].
Hydroxide ions can also damage the polymer matrix of the membrane. It was shown in [bib_ref] Alkali attack on cation-exchange membranes with polyvinyl chloride backing and binder: Comparison..., Doi [/bib_ref] [bib_ref] Alkali attack on anion exchange membranes with PVC backing and binder: II..., Doi [/bib_ref] that OHions reacts with polyvinyl chloride, which in the studied cases was represented by a reinforcing component of Neosepta AMX anion exchange membrane [fig_ref] Figure 6: A color change trend of commercial AMX during the alkali immersion test [/fig_ref] and Neosepta CMX cation exchange membrane. The reaction splat off hydrogen chloride from the polymer and generated polyenes. These compounds lower the Young's modulus of the membrane and increase its water content. In the aforementioned works it was shown that PVC of the Neosepta AMX membrane enters the cleavage reaction in much milder conditions compared to the PVC of the Neosepta CMX membrane, which is associated with the Donnan exclusion of hydroxide ions from the matrix of the cation exchange membrane. It was reported in [bib_ref] Constructing the basal nanofibers suit of layer-by-layer self-assembly membranes as anion exchange..., Shen [/bib_ref] that OHions are able to interact with the layer-by-layer assembled membrane, the layers of which were anchored by hydrogen bonds. The interaction led to the destruction of the layered structure of the membrane, and it was proposed to stabilize the layer-by-layer membrane by sandwiching it between two electrospun membranes.
Membrane poisoning is close to fouling in an aspect that some substances get retained by the membrane; however, this term usually refers to the case when the substance does not form continuous layer or particles of sediments but form complexes with ion exchange groups or get absorbed by the membrane matrix instead. A typical example of these interaction is the blockage of ion exchange groups by polyvalent counterions [bib_ref] Poisoning and Sign Reversal of Permselective Membranes, Ko"rösy [/bib_ref] [bib_ref] Electrodialysis of concentrated brines: Effects of multivalent cations, Severin [/bib_ref].
## Physical interactions
In some cases, the components of the treated solution can damage the membrane structure even without chemical interactions.
One of the mechanisms for this is the stretching of the membrane matrix by ions with bulky hydration shells. Long-term electrodialysis of individual solutions containing strongly hydrated anions (NaH2PO4, and KC4H5O6 (potassium hydrotartrate)) and less hydrated inorganic anion (NaCl and NH4Cl) with Neosepta AMX and Neosepta AMX-Sb It was reported in [bib_ref] Constructing the basal nanofibers suit of layer-by-layer self-assembly membranes as anion exchange..., Shen [/bib_ref] that OH − ions are able to interact with the layer-by-layer assembled membrane, the layers of which were anchored by hydrogen bonds. The interaction led to the destruction of the layered structure of the membrane, and it was proposed to stabilize the layer-by-layer membrane by sandwiching it between two electrospun membranes.
Membrane poisoning is close to fouling in an aspect that some substances get retained by the membrane; however, this term usually refers to the case when the substance does not form continuous layer or particles of sediments but form complexes with ion exchange groups or get absorbed by the membrane matrix instead. A typical example of these interaction is the blockage of ion exchange groups by polyvalent counterions [bib_ref] Poisoning and Sign Reversal of Permselective Membranes, Ko"rösy [/bib_ref] [bib_ref] Electrodialysis of concentrated brines: Effects of multivalent cations, Severin [/bib_ref].
## Physical interactions
In some cases, the components of the treated solution can damage the membrane structure even without chemical interactions.
One of the mechanisms for this is the stretching of the membrane matrix by ions with bulky hydration shells. Long-term electrodialysis of individual solutions containing strongly hydrated anions (NaH 2 PO 4 , and KC 4 H 5 O 6 (potassium hydrotartrate)) and less hydrated inorganic anion (NaCl and NH 4 Cl) with Neosepta AMX and Neosepta AMX-Sb anion exchange membranes were studied in [bib_ref] Evolution of current-voltage characteristics and surface morphology of homogeneous anion-exchange membranes during..., Rybalkina [/bib_ref]. It was shown that after 300 h of electrodialysis the experimental limiting current increased by 33, 90 and 128% in the series NaCl < NaH 2 PO 4 < KHT, respectively, and the authors attributed the growth to the destruction of C-C bonds of the ion exchange material, which occurs as a result of stretching the membrane matrix by strongly hydrated ions. The work [bib_ref] Characterization of an anion-exchange membrane subjected to phosphate and sulfate separation by..., Rotta [/bib_ref] also reports the degradation of anion exchange membranes with the loss of their functional groups and the formation of cavities on the membrane surface upon treatment of a solution containing phosphate and sulfate ions [fig_ref] Figure 7: Scanning electron microscopy images of the surface of [/fig_ref]. The second reason for degradation without chemical interactions is an increase in the ionic strength of the solution, which in turn suppresses the electrostatic interactions between the components of the membrane, for example, between the substrate membrane and selective layers. In [bib_ref] Mechanistic insights into the degradation of monovalent selective ion exchange membrane towards..., Ying [/bib_ref] , the devolution of the properties of a monovalent selective membrane during electrodialysis of brines from Llullaillaco Salt Lake (Argentina), in which the total concentration of dissolved solids (409 g/L TDS) was very high, was studied. The investigation demonstrated the degradation of the surface nanostructure and charge functionalization, caused, according to the authors, by a decrease in the electrostatic interaction between the polyethylenimine active layer and the polystyrene-divinylbenzene substrate due to high ionic strength of the treated solution and resulting in a decrease in monovalent selectivity from 16.6 to 5.67 was shown. anion exchange membranes were studied in [bib_ref] Evolution of current-voltage characteristics and surface morphology of homogeneous anion-exchange membranes during..., Rybalkina [/bib_ref]. It was shown that after 300 h of electrodialysis the experimental limiting current increased by 33, 90 and 128% in the series NaCl < NaH2PO4 < KHT, respectively, and the authors attributed the growth to the destruction of C-C bonds of the ion exchange material, which occurs as a result of stretching the membrane matrix by strongly hydrated ions. The work [bib_ref] Characterization of an anion-exchange membrane subjected to phosphate and sulfate separation by..., Rotta [/bib_ref] also reports the degradation of anion exchange membranes with the loss of their functional groups and the formation of cavities on the membrane surface upon treatment of a solution containing phosphate and sulfate ions [fig_ref] Figure 7: Scanning electron microscopy images of the surface of [/fig_ref]. The second reason for degradation without chemical interactions is an increase in the ionic strength of the solution, which in turn suppresses the electrostatic interactions between the components of the membrane, for example, between the substrate membrane and selective layers. In [bib_ref] Mechanistic insights into the degradation of monovalent selective ion exchange membrane towards..., Ying [/bib_ref] , the devolution of the properties of a monovalent selective membrane during electrodialysis of brines from Llullaillaco Salt Lake (Argentina), in which the total concentration of dissolved solids (409 g/L TDS) was very high, was studied. The investigation demonstrated the degradation of the surface nanostructure and charge functionalization, caused, according to the authors, by a decrease in the electrostatic interaction between the polyethylenimine active layer and the polystyrene-divinylbenzene substrate due to high ionic strength of the treated solution and resulting in a decrease in monovalent selectivity from 16.6 to 5.67 was shown.
## Conclusion to mechanisms of degradation
Ion exchange membrane stability may be compromised not only by reversible and partially reversible mechanisms such as fouling but also by other mechanisms that damage the integrity of the membrane and hence are irreversible. This problem is more frequently considered in relation to membrane electrolysis, fuel cells and redox flow batteries where harsher conditions such as presence of active oxidizing species, heating and low humidity are present, but in some conditions these problems are encountered in electrodialysis as well, for example in bipolar membrane electrodialysis and at overlimiting current modes of monopolar membrane electrodialysis, where the membrane itself becomes the source of OH − ions.
The problem of damage to membrane structure has further importance since it follows from the reviewed publications that the processes of membrane degradation are interrelated and can be mutually enhanced. The example of such enhancement is shown in [fig_ref] Figure 8: Example of processes of membrane degradation enhancing each other in a cycle [/fig_ref].
Hence effective prevention of one type of membrane degradation requires the control over other mechanisms of membrane degradation as well.
## Conclusion to mechanisms of degradation
Ion exchange membrane stability may be compromised not only by reversible and partially reversible mechanisms such as fouling but also by other mechanisms that damage the integrity of the membrane and hence are irreversible. This problem is more frequently considered in relation to membrane electrolysis, fuel cells and redox flow batteries where harsher conditions such as presence of active oxidizing species, heating and low humidity are present, but in some conditions these problems are encountered in electrodialysis as well, for example in bipolar membrane electrodialysis and at overlimiting current modes of monopolar membrane electrodialysis, where the membrane itself becomes the source of OHions.
The problem of damage to membrane structure has further importance since it follows from the reviewed publications that the processes of membrane degradation are interrelated and can be mutually enhanced. The example of such enhancement is shown in [fig_ref] Figure 8: Example of processes of membrane degradation enhancing each other in a cycle [/fig_ref]. Hence effective prevention of one type of membrane degradation requires the control over other mechanisms of membrane degradation as well.
## Approaches to prevention of degradation
The methods of prevention or reversal of membrane degradation are even more numerous than the mechanisms of membrane fouling, and the new techniques are constantly proposed. Enhancement of mechanisms of degradation of ion exchange membranes by the action of other mechanisms calls for techniques that can prevent or reverse several mechanisms. Several possible approaches are discussed in short below.
## Pretreatment of solution
If simultaneous presence of several fouling agents presents the biggest problem, then the pretreatment of solutions that would remove one of these components might be the
## Approaches to prevention of degradation
The methods of prevention or reversal of membrane degradation are even more numerous than the mechanisms of membrane fouling, and the new techniques are constantly proposed. Enhancement of mechanisms of degradation of ion exchange membranes by the action of other mechanisms calls for techniques that can prevent or reverse several mechanisms. Several possible approaches are discussed in short below.
## Pretreatment of solution
If simultaneous presence of several fouling agents presents the biggest problem, then the pretreatment of solutions that would remove one of these components might be the key. For example, calcium might be removed from the solutions using pellet reactor pretreatment [bib_ref] Pellet reactor pretreatment: A feasible method to reduce scaling in bipolar membrane..., Tran [/bib_ref].
Pretreatment is usually carried out through dosing of chemicals [bib_ref] Acid precipitation coupled electrodialysis to improve separation of chloride and organics in..., Li [/bib_ref] , through adsorption on some substrate [bib_ref] A novel bio-flocculation combined with electrodialysis process: Efficient removal of pollutants and..., Wang [/bib_ref] [bib_ref] High-efficiency recovery of 1-ethyl-3-methylimidazolium acetate for sugarcane bagasse pretreatment with industrialized technology, Liang [/bib_ref] , through centrifugation or introduction of filtration steps [bib_ref] Investigation of electrodialysis configurations for vinasse desalting and potassium recovery, Barros [/bib_ref] [bib_ref] Deacidification of passion fruit juice by electrodialysis with bipolar membrane after different..., Vera [/bib_ref] [bib_ref] Effective separation of bio-based alpha-ketoglutaric acid from post-fermentation broth using bipolar membrane..., Szczygiełda [/bib_ref]. Addition of steps to treatment procedure adds costs. Dosing of chemicals limits the ecological advantages of electrodialysis technique. In food industry filtration pretreatment may remove valuable compounds from the treated products, which, however, might be reintroduced at later stages [bib_ref] Deacidification of passion fruit juice by electrodialysis with bipolar membrane after different..., Vera [/bib_ref].
Despite the disadvantages discussed above, in some cases the pretreatment is currently absolutely necessary, for example, in wastewater valorization by electrodialysis where high organic matter load causes high probability of fouling [bib_ref] A novel bio-flocculation combined with electrodialysis process: Efficient removal of pollutants and..., Wang [/bib_ref] [fig_ref] Figure 9: SEM micrographs and EDS [/fig_ref] or other processes with high organic load [bib_ref] Investigation of electrodialysis configurations for vinasse desalting and potassium recovery, Barros [/bib_ref]. sorption on some substrate [bib_ref] A novel bio-flocculation combined with electrodialysis process: Efficient removal of pollutants and..., Wang [/bib_ref] [bib_ref] High-efficiency recovery of 1-ethyl-3-methylimidazolium acetate for sugarcane bagasse pretreatment with industrialized technology, Liang [/bib_ref] , through centrifugation or introduction of filtration steps [bib_ref] Investigation of electrodialysis configurations for vinasse desalting and potassium recovery, Barros [/bib_ref] [bib_ref] Deacidification of passion fruit juice by electrodialysis with bipolar membrane after different..., Vera [/bib_ref] [bib_ref] Effective separation of bio-based alpha-ketoglutaric acid from post-fermentation broth using bipolar membrane..., Szczygiełda [/bib_ref]. Addition of steps to treatment procedure adds costs. Dosing of chemicals limits the ecological advantages of electrodialysis technique. In food industry filtration pretreatment may remove valuable compounds from the treated products, which, however, might be reintroduced at later stages [bib_ref] Deacidification of passion fruit juice by electrodialysis with bipolar membrane after different..., Vera [/bib_ref].
Despite the disadvantages discussed above, in some cases the pretreatment is currently absolutely necessary, for example, in wastewater valorization by electrodialysis where high organic matter load causes high probability of fouling [bib_ref] A novel bio-flocculation combined with electrodialysis process: Efficient removal of pollutants and..., Wang [/bib_ref] [fig_ref] Figure 9: SEM micrographs and EDS [/fig_ref] or other processes with high organic load [bib_ref] Investigation of electrodialysis configurations for vinasse desalting and potassium recovery, Barros [/bib_ref].
## Optimization of electric current modes
The first industrial application of electrodialysis dates back to 1954, but the membranes were initially subject to severe fouling, which required the process to be stopped to flush the membranes, which led to the formation of secondary wastewater. The situation changed with the development of electrodialysis reversal, in which the polarity of the electrodes was periodically changed, and the hydraulic regime was simultaneously changed, as a result of which the former desalination chambers became concentration chambers, and vice versa. During electrodialysis reversal, some foulants are removed from the membrane surface and pores; however, some contaminants may be too strongly bound to the membrane surface or form stable fouling layers that will not "purge" from the membrane after polarity reversal process [bib_ref] Fouling of ion exchange membranes used in the electrodialysis reversal advanced water..., Hansima [/bib_ref]. This prompted the search for new ways to prevent and to reverse fouling.
Not only polarity reversal can reduce the intensity of fouling, but also the use of pulsed electric fields. It was shown in [bib_ref] Impact of pulsed electric field on electrodialysis process performance and membrane fouling..., Cifuentes-Araya [/bib_ref] [bib_ref] Effect of pulsed electric field on the electrodialysis performance of phosphate-containing solutions, Rybalkina [/bib_ref] that the use of a pulsed electric field can significantly reduce precipitation [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref] due to the reduced concentration polarization due to relaxation of concentration profiles during the pause interval.
## Optimization of electric current modes
The first industrial application of electrodialysis dates back to 1954, but the membranes were initially subject to severe fouling, which required the process to be stopped to flush the membranes, which led to the formation of secondary wastewater. The situation changed with the development of electrodialysis reversal, in which the polarity of the electrodes was periodically changed, and the hydraulic regime was simultaneously changed, as a result of which the former desalination chambers became concentration chambers, and vice versa. During electrodialysis reversal, some foulants are removed from the membrane surface and pores; however, some contaminants may be too strongly bound to the membrane surface or form stable fouling layers that will not "purge" from the membrane after polarity reversal process [bib_ref] Fouling of ion exchange membranes used in the electrodialysis reversal advanced water..., Hansima [/bib_ref]. This prompted the search for new ways to prevent and to reverse fouling.
Not only polarity reversal can reduce the intensity of fouling, but also the use of pulsed electric fields. It was shown in [bib_ref] Impact of pulsed electric field on electrodialysis process performance and membrane fouling..., Cifuentes-Araya [/bib_ref] [bib_ref] Effect of pulsed electric field on the electrodialysis performance of phosphate-containing solutions, Rybalkina [/bib_ref] that the use of a pulsed electric field can significantly reduce precipitation [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref] due to the reduced concentration polarization due to relaxation of concentration profiles during the pause interval.
## Tailoring hydrodynamic conditions
One of the most powerful tools for prevention of multiple mechanisms of membrane degradation is tailoring of the hydrodynamic conditions [bib_ref] Profiled Ion Exchange Membranes: A Comprehensible Review, Pawlowski [/bib_ref]. The solution flow rate and hydrodynamics in general determine concentration polarization, and with it such significant processes as generation of H + and OHions and fouling. In turn, the hydrodynamics
## Tailoring hydrodynamic conditions
One of the most powerful tools for prevention of multiple mechanisms of membrane degradation is tailoring of the hydrodynamic conditions [bib_ref] Profiled Ion Exchange Membranes: A Comprehensible Review, Pawlowski [/bib_ref]. The solution flow rate and hydrodynamics in general determine concentration polarization, and with it such significant processes as generation of H + and OH − ions and fouling. In turn, the hydrodynamics is determined by the shape of the channel and by the presence of spacer [bib_ref] Comparison of different ED stack conceptions when applied for drinking water production..., Larchet [/bib_ref]. As expected, the enhancement of turbulence within the chambers of electrodialysis apparatuses helps washing out the deposits [bib_ref] Electrodialysis of raw whey and whey fractionated by reverse osmosis and ultrafiltration, Johnson [/bib_ref] [bib_ref] Overview on the hydrodynamic conditions found in industrial systems and its impact..., Fernandes [/bib_ref] and also removing the H + and OH − ions produced in catalytic generation reaction. Spacers of advanced geometry [bib_ref] Revised spacer design to improve hydrodynamics and anti-fouling behavior in reverse electrodialysis..., He [/bib_ref] or novel composition (e.g., ion-conductive spacers [bib_ref] Ion conductive spacers for increased power generation in reverse electrodialysis, Długołęcki [/bib_ref] are used to promote the development of turbulence.
However, nonconductive spacers have a disadvantage of so-called shadow effect on the membrane and on the solution-the covering of the active area of the membrane and making ions transport in tortuous paths, respectively [bib_ref] Effect of spacer geometry on membrane and solution compartment resistances in reverse..., Mehdizadeh [/bib_ref]. Covering of the active area can be avoided by the transition from the flat membranes separated by a spacer to a profiled membranes in an empty channel [bib_ref] Profiled Ion-Exchange Membranes for Reverse and Conventional Electrodialysis, Loza [/bib_ref] (the intermembrane distance is maintained by the alignment of the profile elements [bib_ref] Power generation using profiled membranes in reverse electrodialysis, Vermaas [/bib_ref] , which is one of the challenges of use of profiled membranes). For reverse electrodialysis, it is shown that the transition from spacers to profiled membranes reduces fouling [bib_ref] Fouling in reverse electrodialysis under natural conditions, Vermaas [/bib_ref] [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref].
Membranes 2023, 13, x FOR PEER REVIEW 12 of 19 [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref]. SEM images of cation and anion exchange membranes and profiled plastics. The images in the upper row were in contact with river water, whereas the images in the lower row where in contact with seawater. The four images at the left side are obtained from flat membranes (stack 1, river water side and seawater side), the four images at the right side are obtained from profiled membranes (stack 2) and profiled plastics (stack 3). Most images were saved at a magnification of 5000×, only the flat (cation exchange membrane) in contact with seawater was saved at 10,000×, to see the spherical deposition. Remnants of diatoms, bacteria and spherical deposition are indicated by arrows. Reprinted with permission from Ref. [bib_ref] Fouling in reverse electrodialysis under natural conditions, Vermaas [/bib_ref]. Copyright © 2012 Elsevier Ltd.
## Surface modification of membrane
One of the methods of suppression of undesirable phenomena in membrane system is the modification of the membrane surface.
In [bib_ref] Modification of anion exchange membranes by oxidation of selected chemical sites for..., Kusumoto [/bib_ref] , to prevent fouling of anion-exchange membranes by anionic surfactants, carboxyl groups were created on the membrane surface by oxidation treatment. In the dissociated state the carboxyl groups possess a charge, the sign of which is the same as that of anionic surfactants and opposite to that of the polar groups in the membrane bulk. Thus, an increase in the resistance to fouling of the membrane modified with polar groups, the sign of charge of which is opposite to the sign of charge of the polar groups in the membrane bulk, is shown.
The deposition of a bifunctional polymer containing quaternary ammonium bases on the surface of an anion exchange membrane to reduce the generation of H + and OHions was shown in [bib_ref] Stability of properties of a modified anion-exchange membrane obtained by treating the..., Butylskii [/bib_ref]. Electrochemical impedance spectroscopy showed that the modification is stable for 50 h, after which the separation of the modifier begins, which in turn enhances the generation of H + and OHions. Thus, an increase in the resistance to fouling [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref]. SEM images of cation and anion exchange membranes and profiled plastics. The images in the upper row were in contact with river water, whereas the images in the lower row where in contact with seawater. The four images at the left side are obtained from flat membranes (stack 1, river water side and seawater side), the four images at the right side are obtained from profiled membranes (stack 2) and profiled plastics (stack 3). Most images were saved at a magnification of 5000×, only the flat (cation exchange membrane) in contact with seawater was saved at 10,000×, to see the spherical deposition. Remnants of diatoms, bacteria and spherical deposition are indicated by arrows. Reprinted with permission from Ref. [bib_ref] Fouling in reverse electrodialysis under natural conditions, Vermaas [/bib_ref]. Copyright © 2012 Elsevier Ltd.
## Surface modification of membrane
One of the methods of suppression of undesirable phenomena in membrane system is the modification of the membrane surface.
In [bib_ref] Modification of anion exchange membranes by oxidation of selected chemical sites for..., Kusumoto [/bib_ref] , to prevent fouling of anion-exchange membranes by anionic surfactants, carboxyl groups were created on the membrane surface by oxidation treatment. In the dissociated state the carboxyl groups possess a charge, the sign of which is the same as that of anionic surfactants and opposite to that of the polar groups in the membrane bulk. Thus, an increase in the resistance to fouling of the membrane modified with polar groups, the sign of charge of which is opposite to the sign of charge of the polar groups in the membrane bulk, is shown.
The deposition of a bifunctional polymer containing quaternary ammonium bases on the surface of an anion exchange membrane to reduce the generation of H + and OH − ions was shown in [bib_ref] Stability of properties of a modified anion-exchange membrane obtained by treating the..., Butylskii [/bib_ref]. Electrochemical impedance spectroscopy showed that the modification is stable for 50 h, after which the separation of the modifier begins, which in turn enhances the generation of H + and OH − ions. Thus, an increase in the resistance to fouling of the membrane modified with polar groups sign of charge of which is the same as the sign of charge of the polar groups within the membrane bulk is shown.
Finally, in [bib_ref] Simultaneous improvement of the monovalent anion selectivity and antifouling properties of an..., Mulyati [/bib_ref] , an increase in the resistance of a membrane layer-by-layer-coated with polymers bearing polar groups with alternating signs of charges (in this case, it was polyallylamine and sodium polystyrene sulfonate) to fouling by a sodium dodecylbenzene sulfonate was shown [fig_ref] Figure 1: Hydraulic scheme of the electrodialysis setup [/fig_ref]. As an explanation, increased hydrophilicity and electrostatic interactions with the negatively charged membrane surface (in the case of membranes whose top layer was sodium polystyrene sulfonate) have been proposed. In addition, the intensity of fouling increases with increasing surface roughness [bib_ref] Characterization of protein, peptide and amino acid fouling on ion-exchange and filtration..., Suwal [/bib_ref] , which suggests that membrane modifications aimed at reduction of surface roughness will also suppress membrane fouling (apparently this mechanism played a role in increase of the scaling resistance of heterogeneous membranes after deposition of homogenizing layer [bib_ref] Mitigation of membrane scaling in electrodialysis by electroconvection enhancement, pH adjustment and..., Andreeva [/bib_ref].
## Crosslinking
It has been shown that the modification with the introduction of a hydrophobic crosslinking agent into the anion exchange membrane makes it possible to simultaneously achieve high electrical conductivity and a low swelling ratio [bib_ref] Dual hydrophobic modifications toward anion exchange membranes with both high ion conductivity..., Hu [/bib_ref]. On the other hand, for other processes, an increase in the resistance to fouling after the introduction of a hydrophilic crosslink was shown in [bib_ref] MXene (Ti3C2Tx)/cellulose acetate mixed-matrix membrane enhances fouling resistance and rejection in the..., Azam [/bib_ref] , where it was also found that the membrane with the highest fraction of the introduced crosslinking agent possesses the highest resistance to fouling. Hence, it might be concluded that increased crosslinking improves the fouling resistance regardless of composition of crosslinking agent. However, crosslinking shifts the balance of the selectivity-permeability tradeoff to the selectivity side, and the conductivity of such membranes decreases. In addition, the intensity of fouling increases with increasing surface roughness [bib_ref] Characterization of protein, peptide and amino acid fouling on ion-exchange and filtration..., Suwal [/bib_ref] , which suggests that membrane modifications aimed at reduction of surface roughness will also suppress membrane fouling (apparently this mechanism played a role in increase of the scaling resistance of heterogeneous membranes after deposition of homogenizing layer [bib_ref] Mitigation of membrane scaling in electrodialysis by electroconvection enhancement, pH adjustment and..., Andreeva [/bib_ref].
## Membrane regeneration
## Crosslinking
It has been shown that the modification with the introduction of a hydrophobic crosslinking agent into the anion exchange membrane makes it possible to simultaneously achieve high electrical conductivity and a low swelling ratio [bib_ref] Dual hydrophobic modifications toward anion exchange membranes with both high ion conductivity..., Hu [/bib_ref]. On the other hand, for other processes, an increase in the resistance to fouling after the introduction of a hydrophilic crosslink was shown in [bib_ref] MXene (Ti3C2Tx)/cellulose acetate mixed-matrix membrane enhances fouling resistance and rejection in the..., Azam [/bib_ref] , where it was also found that the membrane with the highest fraction of the introduced crosslinking agent possesses the highest resistance to fouling. Hence, it might be concluded that increased crosslinking improves the fouling resistance regardless of composition of crosslinking agent. However, crosslinking shifts the balance of the selectivity-permeability tradeoff to the selectivity side, and the conductivity of such membranes decreases.
## Membrane regeneration
Regeneration procedures is a way to restore fouled or poisoned membrane to their pristine state. The chemicals used for regeneration tend to depend on the nature of the fouling/poisoning agent. It is understandable that the scaling by CaCO 3 can be reduced by HCl treatment [bib_ref] Mechanisms of chemical cleaning of ion exchange membranes: A case study of..., Guo [/bib_ref] and the membranes that treated cyanide-free brass electrodeposition effluents that contain zinc and copper are regenerated with NaOH solutions [bib_ref] A three-stage chemical cleaning of ion-exchange membranes used in the treatment by..., Barros [/bib_ref] (it is also noteworthy that the article reports that increasing concentration of alkali by itself causes degradation of anion exchange membranes). Article [bib_ref] Mechanisms of chemical cleaning of ion exchange membranes: A case study of..., Guo [/bib_ref] also reports that NaOH was more efficient at removing oils from both anion exchange membrane and from cation exchange membrane.
The organic and colloid fouling is usually cleaned by concentrated solution of salts (to disrupt the electrostatic interactions between the molecules of foulant and facilitate their removal) or by alcohols or their mixture with water. The article [bib_ref] Cleaning of cation-exchange membranes used in electrodialysis for food industry by chemical..., Bdiri [/bib_ref] explored methods for cleaning cation exchange membranes used in the food industry that included soaking into one of the following solutions: NaCl with a concentration of 35 g/L, reconstituted seawater or a water-ethanol mixture, and showed that the best results are demonstrated by cleaning with the water-ethanol mixture, but with this treatment the fraction of electroneutral solution within the membrane increases with time.
Additionally, the membranes operating in food industry treating the solutions with high content of nutrients may require antimicrobial treatment in addition to the fouling control. It can be achieved by cleaning with sodium hypochlorite, but the prolonged contact with it also causes the degradation of the membrane [bib_ref] Structure and properties of heterogeneous and homogeneous ion-exchange membranes subjected to ageing..., Garcia-Vasquez [/bib_ref].
## Conclusions to prevention of membrane degradation and membrane cleaning
Prevention and control of membrane degradation is one of the most studied topics in electrodialysis and other membrane technologies with a multitude of suggested mechanisms. The perfect solution that would allow completely negating fouling (still) does not exist, and it should be noted that several methods aimed at prevention of degradation cause the other form of membrane degradation themselves, which is especially noticeable in the case of alkali cleaning of anion exchange membranes. Damage to the membrane properties during cleaning and the demand for greener production with lower number of added chemicals cause interest in reagentless approaches to prevention of degradation, since the greater attention to electric field modes and hydrodynamics of membrane channel, as well as to novel approaches to membrane modification that would provide a long-lasting result without the need in repeated treatments.
# Conclusions and perspectives
The stability of ion exchange membranes presented a major challenge for electrodialysis since the introduction of this process, and to date, the issues of fouling of the deterioration of the polymer structure of membranes and of the loss of ion exchange groups are an important direction for research and development efforts.
With all the variety of mechanisms of arising processes and their external manifestations that include an increase or decrease in the thickness of the membrane, an increase in the hydrophilicity or hydrophobicity of membranes, the formation of films on the surface of membranes or crystals in their bulk, the result of these processes is very similar. At the end of the service life, the structure of the membrane is disturbed to some extent, the exchange capacity, electrical conductivity and the limiting current of salt ions through the membrane are reduced. The mechanisms for prevention of fouling are very similar and one action can prevent several mechanisms of degradation from occurrence. The same approaches to the prevention of several types of membrane degradation have the advantage of possibility to deal with several mechanisms of membrane devolution simultaneously.
An analysis of publications on the stability of ion exchange membranes in electrodialysis also shows that the majority of the publications are focused on specific water treatment processes that usually operate with commercial membranes (Neosepta AMX and Neosepta CMX manufactured by Astom, MK-40 and MA-41 produced by Shchekinoazot), while it seems promising to study the stability of new membranes, for which improved properties are assumed, in order to assess the retention of improved properties during membrane operation.
Author Contributions: Conceptualization, K.K.; data curation, K.S., A.K. and K.K.; writing-original draft preparation, K.S., A.K. and K.K.; writing-review and editing, K.K.; visualization, K.K.; supervision, K.K.; project administration, K.K.; funding acquisition, K.K. All authors have read and agreed to the published version of the manuscript.
[fig] Figure 1: Hydraulic scheme of the electrodialysis setup: 1-electrodialysis module, 2-desalination (acidification) compartment tank, 3-concentration (alkalization) compartment tank, 4-electrode rinse solution tank, 5-pumps. DC denotes desalination chamber and CC denotes concentration chamber. Reprinted with permission from Ref. [15] under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/ (accessed on 26 December 2022)). Copyright © 2022 by the authors. Licensee MDPI, Basel, Switzerland. [/fig]
[fig] Figure 2: Types of fouling. Reprinted with permission from[28]. Copyright © 2015 Elsevier B.V. [/fig]
[fig] Figure 4: High-resolution scanning electron microscopy images illustrating surface morphology of (a) activated AEM (anion exchange membrane) after immersing the virgin membrane in 0.001 M NaCl for 24 h, (b) AEM fouled by HA (humic acid) and (c) AEM fouled by HA-Ca 2+ . Aggregated HA-Ca2+ foulants on/in AEM pores are labeled, (d) 2% (w/w) NaCl-desorbed, HA-fouled membrane and (e) 2% (w/w) NaCl-desorbed, HA-Ca2+-fouled membrane, (f) 0.03% (w/w) SDS (sodium dodecyl sulfonate)-desorbed, HA-fouled membrane and (g) 0.03% (w/w) SDS-desorbed, HA-Ca 2+ -fouled membrane. Label A presents recovered AEM pores by desorption and label B shows the remaining foulant after desorption. Reprinted with permission from Ref. [41] under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/ (accessed on 26 December 2022)). Copyright © 2021 Published by Elsevier B.V. [/fig]
[fig] Figure 5: Scheme of the mechanisms of degradation of ammonium bases. Reproduced with permission from Ref. [59] under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/ (accessed on 26 December 2022)). Copyright © 2018, the author(s). [/fig]
[fig] Figure 6: A color change trend of commercial AMX during the alkali immersion test. Reprinted with permission from Ref. [63] under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/ (accessed on 26 December 2022)). Copyright © 2018 by the authors. Licensee MDPI, Basel, Switzerland. [/fig]
[fig] Figure 7: Scanning electron microscopy images of the surface of (a,b) original and (c,d) used anionexchange membrane 500× and 1000× of magnification. Reprinted with permission from Ref.[68]. Copyright © 2021 Elsevier B.V. [/fig]
[fig] Figure 8: Example of processes of membrane degradation enhancing each other in a cycle. [/fig]
[fig] Figure 9: SEM micrographs and EDS (energy dispersive spectroscopy) analysis of cationic virgin membrane, membrane fouled with raw vinasse, and membrane fouled with ultrafiltrated vinasse. Reprinted with permission from Ref.[74]. Copyright © 2019 Elsevier B.V. [/fig]
[fig] Funding: This research was funded by the Russian Science Foundation and Kuban Science Foundation, grant number 22-29-20083. Institutional Review Board Statement: Not applicable. Data Availability Statement: The article does not contain new data. [/fig]
|
Molecular genetics of familial nystagmus complicated with cataract and iris anomalies
Purpose: Familial nystagmus complicated with cataract and iris anomalies are genetically heterogeneous, and the pathophysiological mechanisms remain unclear. It is anticipated that mutations in the paired box 6 (PAX6) gene play a major role in pathogenesis of malformations in anterior segment of the eye. In this study, we analyzed PAX6 in a Chinese pedigree of nystagmus, cataract and iris anomalies. This study will provide insights into the genetic basis of this disease. Methods: Complete ophthalmologic examinations were performed on four patients (excluding one dead patient) and four unaffected individuals in this four-generation family. All coding exons of PAX6 were amplified by polymerase chain reaction (PCR), sequenced and compared with reference database. The variations detected were evaluated in available family members as well as 110 normal controls. Possible changes in structure and function of the protein induced by amino acid variance were predicted by bioinformatics analysis. Results: Nystagmus, cataract or iris anomalies were found in all patients of this family, but the severity was different among these patients. A novel missense mutation in PAX6 was identified in all affected individuals but not in asymptomatic members and 110 normal controls. This mutation causes an amino acid substitution of proline to glutamine at position 118 (p.P118Q) of the paired domain of the PAX6 protein. Such a change may cause structural and functional changes of the protein based on bioinformatics analysis. Conclusions: This study added a novel mutation to the existing spectrum of PAX6 mutations, suggesting that a mutation in PAX6 correlated with anterior segment disorders observed in this family.
The anterior segment is the front third of the eye that includes the structures in front of the vitreous humor: the cornea, iris, ciliary body, and lens. Human anterior segment anomalies are generally caused by disturbances early in the development of the eye and have various hereditary patterns [bib_ref] Anterior segment dysgenesis and the developmental glaucomas are complex traits, Gould [/bib_ref] [bib_ref] Axenfeld-Rieger syndrome in the age of molecular genetics* 1, Alward [/bib_ref]. Anterior segment anomalies are complex, continuous spectrum of disorders, including iris anomalies, cataract and nystagmus. Previous studies showed that a variety of anterior segment anomalies may be associated with paired box 6 (PAX6) gene mutations [bib_ref] Mutations at the PAX6 locus are found in heterogeneous anterior segment malformations..., Hanson [/bib_ref] [bib_ref] Getting your Pax straight: Pax proteins in development and disease, Chi [/bib_ref] [bib_ref] A novel PAX6 gene mutation (P118R) in a family with congenital nystagmus..., Sonoda [/bib_ref].
PAX6 (OMIM 607108, GenBank M93650), a member of the paired box gene family, encodes a transcriptional regulator involved in oculogenesis and other developmental processes [bib_ref] The roles of Pax6 in the cornea, retina, and olfactory epithelium of..., Collinson [/bib_ref] [bib_ref] The role of Pax-6 in eye and nasal development, Grindley [/bib_ref]. The human PAX6 gene located on chromosome 11p13, with its role in oculogenesis first demonstrated as deleted or mutated PAX6 was found in patients with iris anomalies [bib_ref] Positional cloning and characterization of a paired box-and homeobox-containing gene from the..., Ton [/bib_ref]. PAX6 spans 22 kb in size and consists of 14 exons and 13 introns [bib_ref] Two independent and interactive DNA-binding subdomains of the Pax6 paired domain are..., Epstein [/bib_ref]. Mutations in PAX6 have been reported in a individuals, including Snellen best-corrected visual acuity test, slit-lamp microscopy examination, intraocular pressure (IOP) measurement, B ultrasonic scan, and fundus examination. Mutation screening and sequence analysis: Genomic DNA was extracted from 200 μl venous blood using a Qiamp Blood Kit (Qiagen, Hilden, Germany). All the procedures were performed according to manufacturer's protocol. DNA integrity was evaluated by 1% agarose gel electrophoresis.
The coding exons 4-13 of PAX6 with intronic flanking sequences were amplified by PCR using previously published primers [bib_ref] Axenfeld-Rieger syndrome in the age of molecular genetics* 1, Alward [/bib_ref] [bib_ref] The Human PAX6 Mutation Database, Brown [/bib_ref] [fig_ref] TABLE 1: PRIMERS USED IN POLYMERASE CHAIN REACTION FOR AMPLIFICATION OF PAX6 [/fig_ref]. The primers were synthesized by Invitrogen Company (Carlsbad, CA). PCRs were performed in a MyCycler thermocycler (Bio-Rad, Hercules, CA) using the following program: initial denaturation at 95 °C for 2 min followed by 35 cycles of 94 °C for 10 s, 51 °C-56 °C for 30 s, and 72 °C for 1 min, and final extension at 72 °C for 5 min.
PCR products were directly sequenced using an ABI 377XL automated DNA sequencer (Applied Biosystems, Foster City, CA). Sequence data were compared pair-wise with the published PAX6 sequence. Mutation was named according to the nomenclature recommended by the Human Genomic Variation Society (HGVS). Bioinformatics analysis: The Clustalw tool was used to align the protein sequences among eight different species. The possible functional impact of an amino acid change was predicted by Polyphen and SIFT. The isoelectric point (pI) and molecular weight of the wild type and mutant protein were analyzed by the Compute pI/Mw tool. The mutation region in 3D structure of PAX6 protein was analyzed by the Cn3D tool.
# Results
## Clinical findings of the examined family:
Four patients in four successive generations of this family were found to have a similar congenital ocular disorder. The segregation of the ocular anomaly is consistent with an autosomal dominant inheritance.
The proband (patient IV:2, a seven-year-old boy, [fig_ref] Figure 1: Pedigree of the Chinese family with nystagmus, cataract, and iris anomalies [/fig_ref] presented with nystagmus, cataract, and iris anomalies (corectopia and coloboma of iris). In addition, this patient also presented with mental retardation and muscle spasms. His corrected visual acuity was 0.1 OD and 0.1 OS. Intraocular pressure (IOP) was 17 mmHg in the right eye and 16.8 mmHg in the left eye. He was noted to have nystagmus after birth. Clinically, he showed rapid jerk nystagmus with poorly developed fixation ability [fig_ref] Figure 2: Anterior segment of proband [/fig_ref]. B-scanning revealed abnormally short axial lengths and deep anterior chambers (data not shown). Fundus details could not be seen, but retinas were attached per ultrasound. The proband's mother (Patient III:5, thirty-four-year-old, [fig_ref] Figure 1: Pedigree of the Chinese family with nystagmus, cataract, and iris anomalies [/fig_ref] was diagnosed with nystagmus, cataract, and mild iris defects (irregular pupils and mild corectopia; [fig_ref] Figure 2: Anterior segment of proband [/fig_ref]. Her visual acuity was 0.1 OD and 0.1 OS, and IOP was 10 mmHg OD and 13 mmHg OS. Fundus details could not be seen.
The proband's grandmother (Patient II:4, fifty-threeyear-old, [fig_ref] Figure 1: Pedigree of the Chinese family with nystagmus, cataract, and iris anomalies [/fig_ref] had mental illness. She became blind ten years ago and had nystagmus, cataract and anomalies of iris (data not shown). The proband's great-grandmother (Patient I:2, eighty-five-year-old, [fig_ref] Figure 1: Pedigree of the Chinese family with nystagmus, cataract, and iris anomalies [/fig_ref] had various illness and was bed-ridden and couldn't be examined by slit-lamp microscopy. She was blind for more than 30 years and had nystagmus, cataract, and iris anomalies (She was examined with direct ophthalmoscope and flashlight). Patient II:5 was deceased in his thirties by accident. His elder sister recalled that his vision was very poor and had nystagmus at his early age.
Ocular abnormalities were not found in four unaffected members examined in this family.
## Pax6 mutation identification and analysis:
A novel heterozygous mutation, c.353 C>A, at codon 118 (CCA to CAA) of exon 6 in PAX6 was identified. This heterozygous mutation was present in all affected individuals whereas none of the unaffected family members and 110 normal control subjects examined carried the mutation [fig_ref] Figure 3: Sequencing results of the PAX6 gene [/fig_ref]. The affected individuals were heterozygous alleles with mutant and wild-type alleles. The unaffected members carried wildtype alleles. Bioinformatics analysis: PAX6 has two DNA-binding domains, a paired domain (PD) and a homeodomain (HD) and this mutation is in a paired domain. PAX6 with the c.353 C>A mutation would result in replacement of proline by glutamine, leading to a change of a hydrophobic amino acid to a hydrophilic amino acid. Proline in this position was found highly conserved for PAX6 by analyzing orthologs from eight different species using Clustalw tool on line [fig_ref] Figure 4: The mutation involved a highly conserved residue [/fig_ref]. The p.P118Q mutation was predicted to be "probably damaging" by Polyphen with a score of 2.736, and "affect protein function" by SIFT with a score of 0.00. The theoretical pI of mutant PAX6 was 9.45 and no change compared with wild type PAX6. The Mw of the mutant PAX6 was slightly increased to 46714.39 Da from the wild type PAX6 of 46683.37 Da. Cn3D tool analysis also showed that p.P118Q mutation was located in the COOH-terminal region of PAX6 paired domain [fig_ref] Figure 5: The paired domain of PAX6 [/fig_ref] [bib_ref] Crystal structure of a paired domain-DNA complex at 2.5 resolution reveals structural..., Xu [/bib_ref].
# Discussion
All patients of this Chinese pedigree had different degree of anterior segment anomalies since childhood. A 7-year-old patient also presented with mental retardation and muscle spasms. The disease in this family showed autosomal dominant inheritance and the main symptoms included nystagmus, cataract, and iris anomalies. The pathophysiological mechanisms underlying this disease remain unclear. Many studies showed that PAX6 is important for development of the eye and central nervous tissues. Dysfunction of PAX6 causes defects in eye development from Drosophila to human, as PAX6 mutations are known to cause eyeless in Drosophila, small-eye phenotype in the mouse, and iris anomalies and Peters' anomaly in humans [bib_ref] Mutations at the PAX6 locus are found in heterogeneous anterior segment malformations..., Hanson [/bib_ref] [bib_ref] Mouse small eye results from mutations in a paired-like homeoboxcontaining gene, Hill [/bib_ref] [bib_ref] Homology of the eyeless gene of Drosophila to the Small eye gene..., Quiring [/bib_ref]. To date, genetic analysis has detected numerous mutations of PAX6 in patients with iris anomalies [bib_ref] Unexpected allelic heterogeneity and spectrum of mutations in Fowler syndrome revealed by..., Lalonde [/bib_ref]. In this study, a novel heterozygous mutation (c.353 C>A, p.P118Q) in PAX6 was identified in a Chinese pedigree. This mutation was identified in all affected patients (I:2, II:4, III:5, and IV:2), but not in unaffected family menbers and normal control subjects. This mutation was predicted to be deleterious by both Polyphen and SIFT with consistent results. Therefore, it was considered that this variation appears to be a causative mutation of the disease in this family.
PAX6 may regulate the expression of other genes during embryogenesis as a transcription factor [bib_ref] Getting your Pax straight: Pax proteins in development and disease, Chi [/bib_ref] [bib_ref] Two independent and interactive DNA-binding subdomains of the Pax6 paired domain are..., Epstein [/bib_ref]. PAX6 expresses two isoforms by alternative splicing, PAX6 (−5a; 422 amino acids) and PAX6 (+5a; 436 amino acids) [bib_ref] Crystal structure of a paired domain-DNA complex at 2.5 resolution reveals structural..., Xu [/bib_ref] [bib_ref] DNA sequence recognition by Pax proteins: bipartite structure of the paired domain..., Czerny [/bib_ref]. The paired domain (PD) and homeodomain (HD) are two DNA-binding domains of PAX6 and the paired domain consists of 128 amino acids and contains two globular subdomains (NH2-subdomain and COOH-subdomain) linked by an extended polypeptide chain [bib_ref] Genomic structure, evolutionary conservation and aniridia mutations in the human PAX6 gene, Glaser [/bib_ref] [bib_ref] Crystal structure of the human Pax6 paired domain¨CDNA complex reveals specific roles..., Xu [/bib_ref]. This novel p.P118Q mutation was in the COOH-terminal region of the paired domain. The p.P118Q mutation turned a hydrophobic amino acid (proline) into a hydrophilic amino acid (glutamine), and since it is located in the hydrophobic core of the COOH-subdomain, it may disrupt the folding and stability of the whole protein [fig_ref] Figure 5: The paired domain of PAX6 [/fig_ref]. The proline at codon 118 is highly conserved in PAX6 paired-domains of all species, including Mus musculus, Rattus norvegicus, Macaca mulatta, Bos taurus, Gorilla gorilla, Gallus gallus, Danio rerio, and Xenopus tropicalis [fig_ref] Figure 4: The mutation involved a highly conserved residue [/fig_ref]. Such a high degree of conservation argues for a functional importance of the relevant amino acid residue.
A similar mutation, p.P118R (c.353 C>G), of PAX6 was previously found in a Japanese family [bib_ref] A novel PAX6 gene mutation (P118R) in a family with congenital nystagmus..., Sonoda [/bib_ref]. This family showed a variety of congenital ocular disorders, including aniridia, congenital nystagmus, minimal displaced pupil, and foveal hypoplasia. The proline to glutamine (p.P118Q) substitution in the COOH-terminal part of the paired domain reported here has a phenotypic outcome different from that seen when the same residue is substituted with arginine (p.P118R). The Human PAX6 Allelic Variant Database showed that at the same codon more than 1 variant missense base-pair mutation leading to production of different amino acids is found. This illustrates how different missense mutations affecting the same codon of PAX6 may result in distinctly different phenotypes [bib_ref] Clinical manifestation of a novel PAX6 mutation Arg128Pro, Bredrup [/bib_ref]. The affected individuals in this family showed congenital ocular disorders, including nystagmus, cataract, and iris anomalies. Besides these ocular symptoms, the proband (IV: 2) also had mental retardation and skeletal muscle spasms. His grandmother (II:4) was diagnosed as a schizophrenic disorder in her early 30s. PAX6 expresses in the developing neural tissues of the brain, such as cerebellum, and plays a role in the development of neural system and the eye [bib_ref] A novel PAX6 gene mutation (P118R) in a family with congenital nystagmus..., Sonoda [/bib_ref] [bib_ref] PAX-6 in development and evolution, Callaerts [/bib_ref]. Some PAX6 mutations may play a critical role in controlling the migration and differentiation of specific neuronal progenitor cells in the brain [bib_ref] PAX6 gene dosage effect in a family with congenital cataracts, aniridia, anophthalmia..., Glaser [/bib_ref]. Graziano et al. [bib_ref] A de novo nonsense mutation of PAX6 gene in a patient with..., Graziano [/bib_ref] reported a de novo nonsense mutation S119R (serine to arginine) of PAX6 in a patient with aniridia, ataxia, and mental retardation. This mutation may be responsible for aniridia, ptosis and cognitive dysfunction. Davis et al. [bib_ref] Pax6 3′ deletion results in aniridia, autism and mental retardation, Davis [/bib_ref] analyzed of a patient with aniridia, autism, and mental retardation and identified a 1.3 Mb deletion of PAX6. Other findings also provided evidence that the role of PAX 6 in brain anomalies associated with iris anomalies [bib_ref] PAX6 mutation in a family with aniridia, congenital ptosis, and mental retardation, Malandrini [/bib_ref] [bib_ref] PAX6 aniridia and interhemispheric brain anomalies, Abouzeid [/bib_ref].
In summary, this study added a novel missense mutation to the existing spectrum of PAX6 mutations in a Chinese family with nystagmus, cataract and iris anomalies. This study provides further evidence that haploinsufficiency of the PAX6 gene causes defects in eye development and it also adds more evidences for genetic diagnosis of these inherited disorders.
[fig] Figure 1: Pedigree of the Chinese family with nystagmus, cataract, and iris anomalies. Filled squares and circles are affected males and females, respectively. Arrowhead indicates proband. The asterisk indicates family members included this study. [/fig]
[fig] Figure 3: Sequencing results of the PAX6 gene. A: a heterozygous C to A transversion at codon 118 in one patient from the family (arrowhead). B: Wild type sequence from an unaffected member. [/fig]
[fig] Figure 2: Anterior segment of proband (IV:2) and the proband's mother (III:5). A and B show iris anomalies, and nasally and superiorly displaced pupil and cataract in both eyes of proband (IV: 2). C and D show mild nasally displaced pupil and cataract of the proband's mother (III:5). [/fig]
[fig] Figure 4: The mutation involved a highly conserved residue. The proline at position 118 is highly conserved for PAX6, which was demonstrated by analysis of orthologs from eight different species. [/fig]
[fig] Figure 5: The paired domain of PAX6. A: Sketch of the PAX6 paired domain-DNA complex. Cylinders represent α helices; arrows represent β strands. Helices 1-6 are labeled; residue numbers indicate termini of the corresponding secondary structure elements. B: Cn3D display for the paired domain of PAX6. The yellow segment represents the mutation region (COOH-terminal region in PAX6 paired domain). [/fig]
[table] TABLE 1: PRIMERS USED IN POLYMERASE CHAIN REACTION FOR AMPLIFICATION OF PAX6. [/table]
|
Azithromycin versus placebo for the treatment of HIV-associated chronic lung disease in children and adolescents (BREATHE trial): study protocol for a randomised controlled trial
Background: Human immunodeficiency virus (HIV)-related chronic lung disease (CLD) among children is associated with substantial morbidity, despite antiretroviral therapy. This may be a consequence of repeated respiratory tract infections and/or dysregulated immune activation that accompanies HIV infection. Macrolides have anti-inflammatory and antimicrobial properties, and we hypothesised that azithromycin would reduce decline in lung function and morbidity through preventing respiratory tract infections and controlling systemic inflammation.Methods/design: We are conducting a multicentre (Malawi and Zimbabwe), double-blind, randomised controlled trial of a 12-month course of weekly azithromycin versus placebo. The primary outcome is the mean change in forced expiratory volume in 1 second (FEV 1 ) z-score at 12 months. Participants are followed up to 18 months to explore the durability of effect. Secondary outcomes are FEV 1 z-score at 18 months, time to death, time to first acute respiratory exacerbation, number of exacerbations, number of hospitalisations, weight for age z-score at 12 and 18 months, number of adverse events, number of malaria episodes, number of bloodstream Salmonella typhi infections and number of gastroenteritis episodes. Participants will be followed up 3-monthly, and lung function will be assessed every 6 months. Laboratory substudies will be done to investigate the impact of azithromycin on systemic inflammation and on development of antimicrobial resistance as well as impact on the nasopharyngeal, lung and gut microbiome. Discussion: The results of this trial will be of clinical relevance because there are no established guidelines on the treatment and management of HIV-associated CLD in children in sub-Saharan Africa, where 80% of the world's HIVinfected children live and where HIV-associated CLD is highly prevalent. Trial registration: ClinicalTrials.gov, NCT02426112. Registered on 21 April 2015.
# Background
Respiratory disease is the most common manifestation of human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) among children in sub-Saharan Africa, accounting for more than 50% of HIV-associated mortality [bib_ref] The challenge of chronic lung disease in HIVinfected children and adolescents, Weber [/bib_ref] [bib_ref] Global considerations in human immunodeficiency virus-associated respiratory disease, Rylance [/bib_ref] [bib_ref] Chest radiography patterns in 75 adolescents with vertically-acquired human immunodeficiency virus (HIV)..., Desai [/bib_ref] [bib_ref] Chronic lung disease in adolescents with delayed diagnosis of vertically acquired HIV..., Ferrand [/bib_ref] [bib_ref] Clinical characteristics and lung function in older children vertically infected with human..., Mwalukomo [/bib_ref] [bib_ref] Chronic lung disease in HIV-infected children established on antiretroviral therapy, Rylance [/bib_ref] [bib_ref] Risk factors for hypoxia and tachypnea among adolescents with vertically-acquired HIV in..., Attia [/bib_ref]. The use of antiretroviral therapy (ART) and co-trimoxazole prophylaxis has contributed to a reduction in the rate of acute respiratory tract infections and mortality among HIV-infected children in both high-and low-resource settings [bib_ref] Post-HAART outcomes in pediatric populations: comparison of resource-limited and developed countries, Peacock-Villada [/bib_ref].
However, studies in recent years in sub-Saharan Africa have demonstrated that about 30% of perinatally HIVinfected older children and adolescents have chronic respiratory symptoms, including chronic cough, reduced exercise tolerance and significantly impaired lung function [bib_ref] Chronic lung disease in adolescents with delayed diagnosis of vertically acquired HIV..., Ferrand [/bib_ref] [bib_ref] Risk factors for hypoxia and tachypnea among adolescents with vertically-acquired HIV in..., Attia [/bib_ref]. In these studies, even participants with pronounced respiratory impairment looked well at rest, and plain radiological abnormalities were subtle. High-resolution computed tomography findings showed predominantly small airway disease consistent with constrictive obliterative bronchiolitis (OB) [bib_ref] Chronic lung disease in adolescents with delayed diagnosis of vertically acquired HIV..., Ferrand [/bib_ref]. Importantly, no association was observed between abnormal lung function, ART use or duration or CD4 count, suggesting that this form of HIV-related chronic lung disease may not be responsive to ART.
OB is a chronic obstructive lung disease that follows a severe insult to the lower respiratory tract, resulting in fibrosis of the small airways [bib_ref] Bronchiolitis obliterans in children, Moonnumakal [/bib_ref]. The most common presentation is the post-infectious variant, closely related to severe viral infection in the first 3 years of life [bib_ref] Post infectious bronchiolitis obliterans in children, Fischer [/bib_ref]. It is also seen in the context of allogeneic haematopoietic stem cell (HSC) and lung transplant recipients as a result of a graft-versus-host reaction [bib_ref] Stem cell transplantation and lung dysfunction, Haddad [/bib_ref] [bib_ref] Bronchiolitis obliterans after allo-SCT: clinical criteria and treatment options, Uhlving [/bib_ref]. HIV is associated with both high incidence of respiratory infections and persistent immune activation despite ART [bib_ref] Soluble markers of inflammation and coagulation but not T-cell activation predict non-AIDS-defining..., Tenorio [/bib_ref] [bib_ref] Brief report: changes in levels of inflammation after antiretroviral treatment during early..., Macatangay [/bib_ref] [bib_ref] Persistent, albeit reduced, chronic inflammation in persons starting antiretroviral therapy in acute..., Sereti [/bib_ref]. Thus, HIV-associated OB may share causal pathways with both post-infectious and post-transplant variants.
Evidence regarding the efficacy of treatment modalities in OB is sparse. Observational studies have shown some improvement of lung function in a small number of children treated with high-dose pulse corticosteroids for post-transplant OB [bib_ref] Bronchiolitis obliterans after allo-SCT: clinical criteria and treatment options, Uhlving [/bib_ref] [bib_ref] High-dose corticosteroid therapy for bronchiolitis obliterans after bone marrow transplantation in children, Ratjen [/bib_ref]. Studies in lung transplantassociated OB have shown a positive effect of azithromycin, whereas a small randomised controlled trial of patients with established HSC-associated OB failed to show an effect [bib_ref] Azithromycin improves lung function in patients with post-lung transplant bronchiolitis obliterans syndrome:..., Kingah [/bib_ref] [bib_ref] Effects of azithromycin in bronchiolitis obliterans syndrome after hematopoietic SCT-a randomized doubleblinded..., Lam [/bib_ref]. Authors of case series including patients with post-infectious and post-transplant OB have reported a benefit of azithromycin on lung function and on rate of exacerbations [bib_ref] Azithromycin for prevention of exacerbations in non-cystic fibrosis bronchiectasis (EMBRACE): a randomised,..., Wong [/bib_ref].
Azithromycin is a macrolide antibiotic with bacteriostatic activity against the most common respiratory bacterial pathogens. However, it also has a robust immunomodulatory effect resulting in decreased production of proinflammatory cytokines in the acute phase and resolution of chronic inflammation in the later phases. Specifically, azithromycin has direct activity on airway epithelial cells to maintain their function and reduce mucus secretion [bib_ref] Effects of azithromycin in bronchiolitis obliterans syndrome after hematopoietic SCT-a randomized doubleblinded..., Lam [/bib_ref] [bib_ref] Immunomodulatory indications of azithromycin in respiratory disease: a concise review for the..., Cramer [/bib_ref] [bib_ref] Azithromycin reduces airway neutrophilia and interleukin-8 in patients with bronchiolitis obliterans syndrome, Verleden [/bib_ref] [bib_ref] Azithromycin ameliorates airway remodeling via inhibiting airway epithelium apoptosis, Liu [/bib_ref]. These characteristics have resulted in the use of azithromycin in the management of a variety of chronic lung diseases, including cystic fibrosis [bib_ref] Macrolide antibiotics for cystic fibrosis, Southern [/bib_ref] , non-cystic fibrosis bronchiectasis [bib_ref] Long-term azithromycin for indigenous children with non-cystic-fibrosis bronchiectasis or chronic suppurative lung..., Valery [/bib_ref] , bronchiolitis obliterans syndrome [bib_ref] Azithromycin improves lung function in patients with post-lung transplant bronchiolitis obliterans syndrome:..., Kingah [/bib_ref] [bib_ref] Post-infectious bronchiolitis obliterans in children: a review of 42 cases, Li [/bib_ref] and chronic obstructive pulmonary disease [bib_ref] Azithromycin maintenance treatment in patients with frequent exacerbations of chronic obstructive pulmonary..., Uzun [/bib_ref]. The intracellular uptake of azithromycin is high, and hepatic excretion is slow, resulting in a long half-life that enables infrequent dosing. We will test the hypothesis that prophylactic azithromycin is effective, through its antimicrobial and anti-inflammatory properties, in preventing worsening of lung function and in reducing exacerbations in children and adolescents receiving ART who have HIVassociated chronic lung disease.
# Methods/design
Study design BREATHE (Bronchopulmonary function in response to azithromycin treatment for chronic lung disease in HIVinfected children) is a two-site, double-blind, placebocontrolled, individually randomised trial in which we intend to enrol 400 perinatally HIV-infected children and adolescents aged 6-19 years with HIV-associated chronic lung disease who have been receiving ART for a minimum of 6 months. Participants will be enrolled from the outpatient HIV clinics in Harare, Zimbabwe, and Blantyre, Malawi.
## Study intervention
Participants are randomly assigned to receive either azithromycin or placebo in a 1:1 ratio. Weight-band azithromycin (10-19.9 kg, 250 mg; 20-29.9 kg, 500 mg; 30-39.9 kg, 750 mg; > 40 kg, 1250 mg/week) or placebo will be given weekly under direct observation by a treatment monitor identified within the family for a total of 12 months.
## Study population
HIV-infected children and adolescents with HIVassociated chronic lung disease attending outpatient HIV clinics in Harare and Blantyre will be enrolled in the trial. Individuals aged 6-19 years will be approached together with their guardians during their routine HIV outpatient visits and provided with information about the study. Once informed consent is obtained for screening, eligibility will be established using a multi-step screening procedure. Chronic lung disease will be established by spirometry (forced expiratory volume in 1 second [FEV 1 ] z-score less than −1.0) with no reversibility (< 12% improvement in FEV 1 after salbutamol 200 μg inhaled using a spacer). Spirometry will be performed using the EasyOne™ spirometer (ndd Medical Technologies Inc., Andover, MA, USA) by trained research staff certified in performing spirometry and following the American Thoracic Society guidelines. Those who meet the criteria for chronic lung disease will undergo further tests, including a urine pregnancy test (for girls who have reached menarche), serum creatinine, alanine aminotransferase (ALT), an electrocardiogram and screening for tuberculosis (TB). For TB screening, we will use the Xpert™ MTB/RIF (Cepheid, Sunnyvale, CA, USA) to test one sputum sample obtained either spontaneously or through induction.
## Inclusion/exclusion criteria
Perinatally HIV-infected children and adolescents aged 6-19 years who have been receiving ART for at least 6 months for HIV-associated chronic lung disease, who have a firm home address and a stable guardian, and with consent from the guardian and assent from the participant (for those aged < 18 years; those aged ≥ 18 years to consent independently) will be eligible for inclusion. Exclusion criteria will include having a condition that may prove fatal during the study period (e.g., malignancy), TB or acute respiratory tract infection at the time of screening, pregnancy or breastfeeding, history of cardiac arrhythmia, a prolonged QTc interval, abnormal creatinine clearance or elevated ALT, known macrolide hypersensitivity, and concomitant use of digoxin and/or fluconazole (or other drugs known to prolong the QTc interval) [fig_ref] Table 1: Inclusion/exclusion criteria History of cholestatic jaundice or hepatic dysfunction associated with previous... [/fig_ref].
## Study procedures
Eligible individuals will be randomised to either azithromycin or placebo. The allocation ratio will be 1:1 by block randomisation with variable-length blocks stratified by site. The randomisation schedule and allocation list will be generated using Stata™ version 14 software (2015 release; StataCorp, College Station, TX, USA) by an independent statistician not connected with the study. The allocation list will be sent directly to the study pharmacists at both trial sites, who will prepare the study medication. Participants and study personnel will therefore be masked to treatment allocation. The pharmacist will be provided with only randomly allocated numbers assigning participants to arm 1 or arm 2, with each number linked to a study number, and therefore the pharmacist will also remain blinded. It will not be possible to have pre-prepared medication packs, owing to changes of dose with weight.
After randomisation, spirometry will be repeated, and cardiopulmonary fitness will be evaluated using the incremental shuttle walk test. Transthoracic echocardiography will be performed by a trained echocardiographer using the SonoSite™ M-Turbo echocardiography system (FUJIFILM SonoSite, Bothell, WA, USA) at the Malawi site and the Mindray™ DC-N6 echocardiography system (Mindray, Shenzhen, China) at the Zimbabwe site to assess right heart function and pulmonary hypertension. The research nurse will take a rectal swab, nasal aspirate, and sputum and blood samples. The blood sample will be used to measure CD4 count with the Pima™ Analyser (Alere, Orlando, FL, USA) and to perform viral load testing (Xpert™ HIV-1 Viral Load; Cepheid). The study drug will be dispensed by the study pharmacist. After the enrolment visit, follow-up visits will be scheduled at 2 weeks and 3-monthly thereafter. Participants will be followed for a minimum of 12 months to ascertain the primary outcome and for a further 6 months (i.e., 6 months following completion of study drug) to investigate durability of effect (if any) of the intervention. Procedures undertaken at the follow-up visits are summarised in [fig_ref] Figure 1: Schedule of trial procedures [/fig_ref].
Participants will be instructed to attend unscheduled visits if they develop acute symptoms. For specific acute symptoms, investigations and management will be as follows: for diarrhoeal episodes (Division of AIDS [DAIDS] severity grade ≥ 3), a rectal swab and Clostridium difficile rapid test will be taken (C. Diff Quick Check Complete™; Alere) and if positive will be treated with metronidazole for 7 days. For acute respiratory exacerbations, sputum and nasal swabs will be taken and will be treated with amoxicillin-clavulanate for 10 days; if there is no improvement, a chest radiograph and culture for TB will be obtained. For febrile episodes, blood cultures, malaria slides and or malaria rapid diagnostic tests and rectal swabs will be taken, and management will be carried out according to national guidelines. For other acute symptoms, investigations and management will be carried out at the discretion of the treating clinician. A nasal swab, rectal swab, and sputum and blood samples will be collected at study visits for measurement of immune activation markers, investigation of antimicrobial resistance and assessment of changes in the gut and respiratory microbiome.
## Risks
Potential risks include an allergic reaction or adverse reaction to the medication or placebo. Examples of potential side effects include nausea, vomiting, diarrhoea and skin reactions. Each of these risks and any other unexpected outcomes will be monitored at study visits. Criteria for stopping the study drug are allergic reaction, QTc prolongation > 500 milliseconds, hepatic toxicity with ALT > 80 IU/L, concomitant use of drugs known to produce QTc prolongation, pregnancy during the course of the study and any adverse reaction with DAIDS severity grade > 3. In case of any serious adverse events, the institutional review boards and the data and monitoring safety board (DSMB) will be notified. Unblinding of treating physicians will be carried out in case of pregnancy or other events after decision by the DSMB.
## Outcome measures
The primary outcome is the mean difference between trial arms in FEV 1 z-scores (generated using Global Lung Function Initiative reference standards) between trial arms at 12 months after initiation of the study drug, adjusted for site and baseline FEV 1 z-score. The secondary outcomes are described in [fig_ref] Table 2: Secondary outcomes [/fig_ref].
Other outcomes will be the effect of azithromycin on (1) diversity and composition of respiratory and gut microbiome of children with HIV-associated chronic lung disease, (2) antibiotic resistance of bacteria colonising the respiratory tract, and (3) biomarkers of systemic inflammation.
## Laboratory procedures
Nasopharyngeal and sputum samples will be frozen and stored at −80°C for later batch processing, which will include routine bacterial culture for respiratory pathogens, antimicrobial susceptibility testing and 16S ribosomal RNA (rRNA) amplicon sequencing (microbiome analysis). Microbiome analysis of stool samples will also be done.
Archived plasma samples (isolated and stored at −80°C) will be used to quantify the levels of soluble markers of immune activation and microbial translocation, including high-sensitivity C-reactive protein, D-dimers, interleukin-6, β 2 -microglobulin and soluble CD14 using multiplex bead assays (Luminex, Austin, TX, USA). Levels of bacterial 16S rRNA will be quantified using qRT-PCR to measure the extent of microbial translocation over time. The levels of these markers will be correlated with FEV 1 zscores and compared between trial arms.
## Data collection and management and analysis plan
Research nurses and assistants collect the data at baseline and follow-up and record it on electronic clinical record forms using Google Nexus™ tablets (Google, Mountain View, CA, USA) running OpenDataKit software. In both countries, data will be uploaded, processed and saved to a Microsoft Access database (Microsoft, Redmond, WA, USA) before being exported to Stata version 14.0 software and merged into a single database backed up on a regular basis. Consistency checks and checks for missing data are performed both at data entry and fortnightly once the database has been merged. Laboratory data will be recorded on paper forms and entered into the database using optical character recognition. Spirometric data will be uploaded directly from the spirometer. Echocardiographic data will be recorded on electronic clinical record forms, and these will be processed and saved to the Microsoft Access database for merging with the main trial database.
The primary analysis will be done on a modified intention-to-treat basis [bib_ref] Analysis of multicentre trials with continuous outcomes: when and how should we..., Kahan [/bib_ref]. Secondary analyses will include a per-protocol analysis which will be defined using all adherence data prior to unblinding the study. Continuous outcomes will be compared between treatment groups, adjusting for site and baseline measures of the outcome and using linear regression to estimate the mean difference and corresponding 95% CI. Time-toevent outcomes will be assessed using Cox proportional hazards regression and graphically displayed using Kaplan-Meier estimates. Between-group comparisons of binary outcomes will be analysed with logistic regression to estimate ORs and 95% CIs. Count data (e.g., number of hospitalisations) will be analysed using Poisson regression to estimate incidence rate ratios and 95% CIs. All analyses will be adjusted for site. Pre-specified effect modification analyses will include site and baseline severity of lung disease.
## Sample size and power
The following assumptions were made to calculate the sample size:
Mean FEV 1 z-score of −2.04 in the control group Patients randomised in equal proportions to the two regimens Up to 25% of participants unassessable owing to loss to follow-up, death or suboptimal spirometric traces No change in FEV 1 z-score in the control arm Difference in the trial arm mean FEV 1 z-scores ranging from 0.15 to 0.3, an effect assumed to be of clinical relevance SD ranging from 0.55 to 0.82 to assess the impact of variability on the difference in mean value the study has power to detect Under these assumptions, a sample size of 400 recruited participants and 300 participants with outcome data (25% unassessable based on previous studies of Zimbabwean children) will enable 80% power to detect a difference in trial arm means ranging between 0.17 and 0.23, an effect size (difference in means/SD) of 0.32 [fig_ref] Figure 2: Sample size calculations [/fig_ref].
## Ethics and regulatory bodies
The trial will be conducted in accordance with the principles of the Declaration of Helsinki, in compliance with the protocol approved by relevant ethics committees, and according to good clinical practice standards. No change in the protocol will be implemented without the prior review and approval of the regulatory authorities, except where it may be necessary to eliminate an immediate hazard to the trial participants. Details concerning the enrolment process can be found in the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist (Additional file 1) [bib_ref] SPIRIT 2013 statement: defining standard protocol items for clinical trials, Chan [/bib_ref].
The trial sponsor is the London School of Hygiene and Tropical Medicine (LSHTM). The trial will be monitored by the Clinical Trials Unit of the LSHTM and externally by the University of Zimbabwe Research
## Publication
Results of this trial will be published on completion in a peer-reviewed scientific journal. The funder will not be involved in the analysis or interpretation of the data. The full anonymised dataset will be made available no longer than 18 months after completion of the trial.
# Discussion
The massive scale-up of ART programmes globally has resulted in a dramatic improvement in survival, such that increasing numbers of children perinatally infected with HIV, many of whom would have died in early childhood without HIV treatment, are now reaching adolescence. Thus, attention needs to be focused on addressing the chronic complications of long-standing HIV infection prevalent among this cohort of older HIV-infected survivors. Studies in recent years have demonstrated a high burden of chronic lung disease in children with HIV in sub-Saharan Africa, even among those receiving ART and virologically suppressed. The urgency with which evidence-based management guidelines are needed is starkly apparent: Nearly one-third of older children have chronic respiratory symptoms, and in the absence of any alternative therapeutic strategies, repeated treatment for presumptive TB is the only treatment offered and often administered. Our group has investigated bronchodilators and short course, high-dose steroids in patients with chronic lung disease who are receiving ART and isoniazid preventive therapy, with no suggestions of benefit (personal communication, T. Mwalukomo). Whereas in the pre-ART era, lymphoid interstitial pneumonitis (LIP) was the most common cause of chronic lung disease, studies in the ART era show that OB is the most likely underlying cause, with LIP being an exceptional finding. OB is a potentially life-threatening condition that can progress to hypoxic respiratory failure and cor pulmonale and may impair lung growth in children.
OB results from small airway inflammation and subsequent fibrosis, and in the context of HIV, it is likely that inflammation is a consequence of both HIV-mediated chronic immune activation, resulting in end-organ damage, as well as repeated infections due to HIV-mediated immunodeficiency. We therefore postulate that azithromycin is a strong candidate as a therapeutic agent for HIV-associated chronic lung disease, given its broadspectrum antibiotic activity and its anti-inflammatory and immunomodulatory properties, as well as demonstrated activity in similar chronic lung diseases and a good safety profile and tolerability. In this trial, we will investigate the effect of azithromycin on lung function as well as on acute exacerbations and other infections, and the trial will provide insight into the pathogenesis of this condition.
## Trial status
The current protocol is version 2.2, dated 21 August 2017. Enrolment started in June 2016 and will continue until June 2018.
## Availability of data and materials
Datasets generated during the course of this study will be available, after analysis, in a public data repository. Before the datasets are made publicly available, all identifiable information will be removed.
Authors' contributions JOO, RAF and ELC conceived of the study and wrote the basic protocol. JOO established the Principal Investigator's group and applied to the GLOBVAC fund on behalf of the consortium. CGM, KK, GM, ELC and HM contributed to the design and development of the study protocol. AMR led the statistical analysis with input from HAW. MPN and SRJ developed the laboratory components of the study. CGM drafted the manuscript with contributions from RAF and KK. TJG and TF were instrumental in the original protocol design and co-funding. All authors contributed to the manuscript, and all authors read and approved the final version.
Ethics approval and consent to participate Written informed consent by the participants' guardians and assent will be obtained from participants aged younger than 18 years of age using an ageappropriate assent form before enrolment. Participants older than 18 years of age will be able to consent independently.
## Consent for publication
Not applicable.
[fig] Figure 1: Schedule of trial procedures. ALT Alanine aminotransferase, ECG Echocardiography, HIV Human immunodeficiency virus, TB Tuberculosis [/fig]
[fig] Figure 2: Sample size calculations. FEV 1 Forced expiratory volume in 1 second Funding This trial is funded by the Global Health and Vaccination Research (GLOBVAC) Programme of the Medical Research Council of Norway. [/fig]
[table] Table 2: Secondary outcomes [/table]
[table] Table 1: Inclusion/exclusion criteria History of cholestatic jaundice or hepatic dysfunction associated with previous use of azithromycin or known hypersensitivity to a macrolide or ketolide drug • Prolonged QTc interval (> 440 milliseconds in males; > 460 milliseconds in females) • Creatinine clearance < 30 ml/minute • ALT > 80 IU/L • Concomitant use of digoxin and/or fluconazole • Lack of understanding of the study procedure or uncooperative behaviour [/table]
|
Λ-enhanced grey molasses on the D2 transition of Rubidium-87 atoms
Laser cooling based on dark states, i.e. states decoupled from light, has proven to be effective to increase the phase-space density of cold trapped atoms. Dark-states cooling requires open atomic transitions, in contrast to the ordinary laser cooling used for example in magneto-optical traps (MOTs), which operate on closed atomic transitions. For alkali atoms, dark-states cooling is therefore commonly operated on the D 1 transition nS 1/2 → nP 1/2 . We show that, for 87 Rb, thanks to the large hyperfine structure separations the use of this transition is not strictly necessary and that "quasi-dark state" cooling is efficient also on the D 2 line, 5S 1/2 → 5P 3/2 . We report temperatures as low as (4.0 ± 0.3) μK and an increase of almost an order of magnitude in the phase space density with respect to ordinary laser sub-Doppler cooling.Providing several orders of magnitude of gain in phase-space density from a room-temperature atomic vapour, laser cooling is essential to almost all quantum gases experiments. Sub-Doppler cooling, i.e. cooling below the limit temperature of two-level atoms, relies on a combination of ac-Stark shifts and optical pumping among Zeeman sublevels 1 . While intense research on the cooling mechanisms has taken place in the '80s and early '90s, the advent of Bose-Einstein condensation in dilute alkali atoms diverted much of the interest of atomic physicists and laser cooling gradually turned into a tool, ordinarily used and partially understood. Recently, interest in the fundamentals of laser cooling has been revived by the demonstration of effective optical schemes for high-resolution imaging of individual atoms 2-8 , as well as direct laser cooling processes towards quantum degeneracy without any evaporative cooling stage 9,10 . Many of these techniques employ open transitions 5,11 . In fact, cooling on open transitions optically pumps atoms in Zeeman dark states thereby reducing the number of spontaneously emitted photons. Such photons impart a randomly directed recoil to the atoms, that limits the lowest attainable temperature, and generate an effective interatomic repulsion, that limits the highest attainable density 12 . In addition, atoms in excited states cause light-induced losses due to fine-structure changing collisions and radiative escape 13 . Such effects are detrimental to ultracold atoms experiments, for example they limit the reachable phase space density (PSD) in magneto-optical traps (MOTs), hindering the subsequent transfer of the atomic sample into optical or magnetic traps.The existence of "dark" states in open transitions F → F′ between ground F and excited F′(≤F) hyperfine manifolds entails the possibility for atoms to decouple from the laser light. Dark states are linear superpositions of |F, m F 〉 Zeeman sublevels, that depend on the local polarisation of the laser fields. Atoms moving at sufficiently low velocities remain adiabatically dark, as the linear superposition adjusts to the slowly varying polarisation. Instead, faster atoms undergo diabatic transitions towards "bright" states. With blue-detuned light, atoms are more likely to decelerate when in bright states, thus they progressively accumulate near the zero-velocity dark state 14,15 . Such cooling has been investigated since the late '90s and is commonly referred to as "grey molasses" since it involves states neither bright nor completely dark16,17. Recently a twist has been added to the picture 11 : with an additional laser frequency tuned on the repumper transition in Λ-configuration, the dark states become a superposition involving both F − 1 and F hyperfine levels, dominated by the F − 1 level when, as usually the case, the intensity of the light on the cooler transition F → F′ is much larger than the one of the repumper transition F − 1 → F′ 18 . It has been shown with 40
With some alkali atoms, such as Li and K, the hyperfine energy separations in the upper level of the D 2 line (nS 1/2 → nP 3/2 ) are of the same order as the natural linewidths, thus the closed transitions F → F + 1 is hardly isolated from the open transitions. For this reason, for such atoms grey molasses are tipically implemented on the D 1 line, with the recent exception of ref. [bib_ref] Sub-Doppler laser cooling of 40 K with Raman gray molasses on the..., Bruce [/bib_ref]. However, other atoms, such as Rb and Cs, feature nP 3/2 hyperfine separations much larger than the natural linewidths. For these atoms it is worth exploring grey molasses on the D 2 transition that is used for the MOT, with the distinct advantage of avoiding the additional laser source needed to implement grey molasses on the D 1 line.
In this work, we characterise sub-Doppler cooling in 87 Rb with blue-detuned light in a wide range of frequencies blue-detuned with respect to the F = 2 → F′ = 2 open transition (grey molasses). We show that in our experiment the grey molasses reduces the final temperatures by a factor of 4 with respect to the bright molasses, with a minimum observed temperature of (4.0 ± 0.3) μK in a sample of ~10 8 atoms. In addition, PSD is increased by an order of magnitude. These results represent an important advancement for the production of quantum degenerate gases, where laser cooling is most often followed by evaporative cooling, which greatly benefits from beginning at high PSD. Furthermore, the method implemented here can be useful in all experiments using 87 Rb as "coolant" species to realize ultracold atomic mixtures by sympathetic cooling [bib_ref] Bose-Einstein Condensation of Potassium Atoms by Sympathetic Cooling, Modugno [/bib_ref] [bib_ref] Tunable Miscibility in a Dual-Species Bose-Einstein Condensate, Papp [/bib_ref] [bib_ref] A double species 23 Na and 87 Rb Bose-Einstein condensate with tunable..., Wang [/bib_ref].
# Results
In this section, after a brief description of the experiment, we report the characterisation of the grey molasses cooling. For more details about the experimental procedure we refer the reader to the Methods.
We load N MOT = 3 × 10 8 atoms in a MOT at 100 μK from a cold atomic beam in typically 7 s. After the MOT loading, we can increase the PSD of the sample adding a bright molasses stage, which gives as best result 1.5 × 10 8 atoms in the F = 2 hyperfine level at a temperature of (16.8 ± 0.7) μK, and a phase space density λ ≡ = . ± . × − PSD n (4 5 0 6) 10 B d , where n is the peak spatial density and λ π = h mk T / 2 dB B the thermal de Broglie wavelength. These values represent our reference for a comparison with the results obtained with the grey molasses.
The number of atoms in the MOT is monitored by their fluorescence emission: once this has reached a fixed value, we switch off the MOT magnetic fields and start the molasses. To assess the efficiency, immediately after the molasses we measure: the number of remaining atoms, the temperature and the size of the sample. To measure the temperature, atoms are let free to expand for a certain time-of-flight (TOF), then we switch on the MOT laser beams and acquire fluorescence images of the atomic cloud on a CCD camera for different TOF values (more details in Methods).
In most laser cooling experiments the cooler and the repumper lights are typically obtained from two distinct, not phase-coherent, laser sources. For 87 Rb usual laser cooling schemes, the cooler light is quasi-resonant on the F = 2 → F′ = 3 transition and the repumper light is resonant on the F = 1 → F′ = 2 (see [fig_ref] Figure 1: Scheme of the energy levels of the D 2 transition in 87... [/fig_ref]. For grey molasses cooling enhanced by Λ-configuration, the phase-coherence between cooler and repumper is necessary to preserve the linear superpositions of F = 1 and F = 2 sublevels which constitute the (quasi-)dark states (see in section "Adiabatic energy levels: dark states"). Indeed, we elucidate the relevance of phase-coherence between the two laser fields on the efficiency of grey molasses cooling by using alternatively a phase-coherent or a phase-incoherent repumper. The phase-coherent repumper is obtained from a sideband of the cooler, coherently generated by an Electro-Optical Modulator (EOM) with a frequency shift Δ RC /(2π) from the carrier. In [fig_ref] Figure 1: Scheme of the energy levels of the D 2 transition in 87... [/fig_ref] we display a schematic representation of 87 Rb levels and the frequencies used in the molasses phase: cooler(repumper) is blue-detuned with respect to the F = 2(1) → F′ = 2 transition, with detuning Δ 22 (Δ R ). Their frequency difference, as set by the radio-frequency driving the EOM, is Δ RC = Δ R − Δ 22 + E hfs /ħ (E hfs /h = 6834.68 MHz is the energy splitting between the two hyperfine states F = 1 and F = 2 of the atomic ground state 5 2 S 1/2 ).
In order to fully characterise and optimise the grey molasses, we individually vary the repumper intensity, the molasses duration and the light detunings, and we measure the temperature (T), the sizes of the cloud (σ x,z ) and the number of atoms remaining after the grey molasses (N). From the measured data, we extract the PSD of the sample, normalized to the value PSD B obtained with the bright molasses.
Repumper intensity. We start by describing the effect of varying the ratio between repumper and cooler intensities I R /I C , which determines the superposition of states composing the dark-state. We fix the molasses duration Δt = 3 ms, the optical intensity to ~6I s for each beam and the detunings Δ Δ = Γ 5 R 22
(I s = 1.67 mW/ cm 2 and Γ = 2π × 6.065 MHz denote the saturation intensity and linewidth of the D 2 transition, respectively). We set the EOM sidebands frequency at Δ RC /(2π) = 6834.6 MHz, close to the hyperfine splitting E hfs /h, thus resembling a Λ-system, and vary their amplitude, hence I R /I C , by adjusting the radio-frequency power driving the EOM. For each value we measure the temperature T, the fraction of remaining atoms N/N MOT and the normalized phase space density PSD/PSD B extracted from the data. Data in [fig_ref] Figure 2: Effect of repumper intensity [/fig_ref] (left) show that .
[formula] I I / 007 R C [/formula]
is sufficient to have both minimum temperature and maximum atom number while the PSD in [fig_ref] Figure 2: Effect of repumper intensity [/fig_ref] (right) saturates for higher values of the repumper intensity fraction. A relatively small value of the repumper intensity could be convenient as it simplifies the requirements on the power of the sidebands produced by the EOM.
We point out that the lowest temperature in the [fig_ref] Figure 2: Effect of repumper intensity [/fig_ref] (left) is larger than the minimum value reported below because these data were taken with imperfect cancellation of the residual magnetic field, a critical step to optimize the performance of the grey molasses. Data presented hereafter were instead obtained, after proper magnetic field compensation, with I R /I C set to 0.08.
## Time duration.
The time duration of the grey molasses is crucial for its efficiency. Ideally, for perfect molasses, a long duration reduces the final temperature. However, due to the absence of any spatial trapping, the density of the sample decreases as the atoms diffuse and expand. As a consequence, if the parameter to be maximized is the PSD, a compromise arises between lower temperature and higher density.
In order to find its optimal value, we vary the molasses duration Δt at constant detunings Δ Δ = Γ 5
[formula] R 22 [/formula]
and Δ RC /(2π) = 6834.6 MHz. [fig_ref] Figure 3: Optimization of molasses duration [/fig_ref] (left) shows that the temperature initially decreases with Δt, reaching its minimum value T min = (8.6 ± 0.4) μK after 3 ms of molasses, then it flattens and slightly increases for longer times (at 10 ms we measure 12 μK). Conversely, we find that longer molasses capture more atoms. Indeed, since we measure the number of atoms via fluorescence imaging (see Methods for more details), we detect only the atoms that at the moment of the imaging are cold enough to remain within the imaging volume, and this number increases when the temperature decreases. This explains the observed trend. To choose the optimal duration, we plot the PSD in [fig_ref] Figure 3: Optimization of molasses duration [/fig_ref] (right). Within the error bars, the PSD is maximum for time durations from Δt = 3 ms to Δt = 7 ms, and the temperature is minimum within the whole range. Therefore, in the following, we fix the duration Δt to 3 ms.
Phase-coherence and detunings. As expected, the detuning of the cooler with respect to the open transition F = 2 → F′ = 2 (and as a consequence the detuning of the repumper with respect to the F = 1 → F′ = 2 transition) considerably influences the efficiency of the grey molasses. In addition, coherence between the two frequencies is fundamental to preserve the superpositions of atomic sublevels composing the dark states. For this reason, in this section we investigate in detail the dependence of the efficiency of grey molasses on the absolute detunings Δ 22 and Δ R both in the case of coherent and incoherent light, keeping their relative detuning Δ RC fixed In [fig_ref] Figure 4: Influence of detuning [/fig_ref] (left) we compare two datasets: one refers to the grey molasses performed using the coherent repumper light (empty points in the graph), while the other is obtained using the repumper light delivered by another distinct laser source (filled points in the graph), i.e. incoherent with respect to the cooler. In both cases, the temperature decreases as we increase the detuning, but the use of the coherent repumper light clearly provides lower temperatures in the whole range of detunings. The trend of PSD/PSD B is reported in the inset.
In [fig_ref] Figure 4: Influence of detuning [/fig_ref] (right) the same comparison is reported for N/N MOT , where, we remind, N MOT is the total number of atoms measured at the end of the MOT phase. Here, both for the case of coherent and incoherent light, we compare the number of remaining atoms in the F = 1 and F = 2 ground hyperfine levels. As a matter of fact, also the knowledge of the final state of the cooled atoms is important, in particular once the atoms are subsequently transferred, and further cooled, in state-dependent traps. Besides, the final state of the atoms crucially depends on the mechanism underlying the grey molasses, and provides an effective probe on the cooling operation.
In the case of coherent repumper light, for small detunings almost 80% of the atoms are cooled in the grey molasses and they mostly occupy the F = 1 hyperfine state. Increasing the detuning, the total number of atoms drastically decreases and their population is equally distributed among the two hyperfine states F = 1 and F = 2. With incoherent repumper light, instead the grey molasses always captures a small fraction of the MOT atoms; also, the relative population in the two hyperfine levels is almost equal for small detunings while for higher ones the atoms are mostly pumped in the F = 2 state. Close to the resonance, the observed trend is consistent with the expected behavior for our experimental parameters, i.e. the fraction of atoms effectively pumped in F = 1 almost reflects the cooler/repumper intensity ratio. The comparison provided here is a stark evidence that coherent evolution within the ground hyperfine manifold enhances the cooling efficiency of the grey molasses and the accumulation of atoms in F = 1 level.
The best compromise between low temperatures and large atom numbers is found looking at the PSD in the inset of [fig_ref] Figure 4: Influence of detuning [/fig_ref] : we report the case of coherent light (empty points) and incoherent (filled points). Actually, in the case of coherent light, the PSD levels to its maximum value in the broad range of detunings from Δ Γ 4 [bib_ref] Gray-molasses cooling of 39 K to a high phase-space density, Salomon [/bib_ref] to Γ 11 , and it is much higher than in the incoherent case within the whole range of detunings.
Raman detuning. In order to optimize the relative detuning between the coherent repumper and the cooler we vary the EOM frequency, for three different values of Δ 22 : 5 Γ, 8 Γ and 12 Γ. In this section, we define the The temperature measured after the grey molasses is reported in [fig_ref] Figure 5: Raman detuning [/fig_ref] (left): for each value of Δ 22 , the minimum temperature is obtained for δ − . Γ 0 01 R , slightly below the Raman condition δ R = 0, similar to what has been observed in earlier experiments [bib_ref] Sub-Doppler cooling of sodium atoms in gray molasses, Colzi [/bib_ref]. Despite a careful compensation of stray magnetic fields, we could not eliminate the residual shift with respect to the Raman condition. Consistently with the data shown in the previous section, we see that the minimum temperature is approximately the same for the three values of Δ 22 . In the inset of [fig_ref] Figure 5: Raman detuning [/fig_ref] we report the number of remaining atoms; we observe that in the whole range of δ R the higher is Δ 22 , the smaller is N/N MOT , in agrement with the trend observed in [fig_ref] Figure 4: Influence of detuning [/fig_ref].
The corresponding values obtained for PSD/PSD B are reported in [fig_ref] Figure 5: Raman detuning [/fig_ref] (right); we observe a clear peak for δ − . Γ 0 01 R , and the maximum value of the PSD is the same within the error bar for the three different Δ 22 values, as shown in the summary [fig_ref] Table 1: Minimum temperature T, fraction of remaining atoms N/N MOT and normalized phase... [/fig_ref].
However, a notable difference arises for positive values of δ R when we compare the datasets for Δ 22 /Γ = 5, 8, 12. Here, as expected, cooling is less efficient 11 and the final temperature features a Fano profile, which however gets lower for larger detunings, to the point that for Δ 22 = 12 Γ it is no longer visible. The reason of this behaviour might lie in the structure of 87 Rb atomic levels: the energy separation between F′ = 2 and F′ = 3 levels is 44 Γ, thus for high values of Δ 22 the upper level might start to play a role. However, it is still unclear why the effect is visible only for positive δ R ; a deeper understanding could further elucidate how grey molasses works in the presence of a richer level structure, but requires additional investigations.
## Adiabatic energy levels: dark states
It was earlier recognized that physical insight into this cooling mechanism is gained through a 1-dimensional model of Λ-enhanced grey molasses [bib_ref] Λ-enhanced sub-Doppler cooling of lithium atoms in D 1 gray molasses, Grier [/bib_ref] [bib_ref] A Novel Scheme for Efficient Cooling below the Photon Recoil Limit, Weidemüller [/bib_ref] , that takes into account the variations in space of the levels of the full Hamiltonian dressed by the laser fields. Thus we calculate here the position-dependent energy levels by numerical diagonalization of the Hamiltonian
[formula] ∑ = | 〉〈 | − Ω +Ω + → + . . ω ω − − −Ĥ E j F m j F m e e ye xe a h c , , , , 2 [( ) ( ) ] (1) jFm jFm R i t C i t ikz i kz R C [/formula]
taking into account the full hyperfine structure of both the lower 5 2 S 1/2 and upper 5 2 P 3/2 electronic levels of the D 2 transition: E jFm denote the energies of the ground (j = 1/2) and excited (j = 3/2) hyperfine manifolds, with the definition E 1/2,1,m = 0;
[formula] Ω ≡ Γ I I /2 R C R C S ( ) ( ) [/formula]
is the repumper (cooler) Rabi frequency in terms of the saturation intensity I S = 1.67 mW/cm 2 and the excited state linewidth Γ/(2π) = 6.065 MHz, → a are the raising operators of atomic levels whose matrix elements are the 6 − j Wigner coefficient and, finally, ω R(C) the repumper (cooler) angular frequency. We consider a configuration with two counter-propagating beams of orthogonal linear polarizations (lin ⊥ lin). Each beam carries the repumper and cooler frequency ω R ,ω C , with Rabi frequencies corresponding to the total intensities used in the experiment in all six beams, namely Ω C = 4.2 Γ, Ω R = 1.2 Γ. First, we neglect the coupling of the cooler (repumper) with the F = 1(2) → F′ transitions, due to very large detuning (~Γ 10 3 ). Then, we apply the unitary transformation
[formula] , ′ =Δ − Δ E m R C 1/2,2, , ω ′ = − E E F m F m R 3/2, , 3 /2, , [/formula]
. [fig_ref] Figure 6: Energy of low-scattering states as a function of position [/fig_ref] shows the position-dependent eigenvalues of H′. For each state |ψ j 〉, the line-thickness encodes the scattering rate γ ψ ψ ′ = Γ〈 | | 〉 P j j e j . As these are the interesting states for the grey molasses mechanism, we plot only states with γ′ Γ < .
/ 05 j , whose population is predominantly in the two ground hyperfine manifolds. It is clear that the low scattering states are mainly in the F = 1 ground level and that F = 2 states are generally broader, as expected from the relative magnitude of the Rabi frequencies, Ω C > Ω R . We also notice that, for negative Raman detuning δ R = −0.1 Γ (see [fig_ref] Figure 6: Energy of low-scattering states as a function of position [/fig_ref] the level configuration favors cooling as several bright states lie at higher energy than the low-scattering (narrow), predominantly F = 1, states. Conversely, for positive Raman detuning δ R = 0.1 Γ (see [fig_ref] Figure 6: Energy of low-scattering states as a function of position [/fig_ref] (right)), the predominantly F = 1 levels are visibly more mixed with the other levels.
We point out that the grey molasses cooling occurs because, thanks to their motion, atoms in dark states still have a finite probability amplitude of undergoing diabatic transition to a different adiabatic dressed state. Quite reasonably, the probability amplitude of these Landau-Zener (LZ) processes is proportional to the atomic velocity and is larger at the locations where the dressed states get closer in energy (avoided crossings). If, following the diabatic transition, an atom ends in a bright state it faces two possibilities: either climbing or sloping down the light-shift potential as it moves away from the LZ location, the former (latter) leading to loss (gain) of kinetic energy, i.e. cooling (heating). Obviously the bright state energies are periodic in space, thus the cumulative variation of kinetic energy vanishes when the atom travels over one period distance. But the lifetime of bright states is limited by optical pumping, so that the average variation of kinetic energy is determined by the dynamics immediately after the LZ transition.
More quantitatively, we calculate the diabatic couplings ~ψ ψ 〈 |∂ 〉 v j z k for any given state |ψ j 〉 to all other states |ψ k 〉, where ∂ z and v denote the gradient and the velocity in our 1D model. The diabatic couplings confirm the cooling scenario described above, showing for example that the lowest-energy, predominantly F = 2, level in [fig_ref] Figure 6: Energy of low-scattering states as a function of position [/fig_ref] is not coupled to the low-scattering, predominantly F = 1, states.
We notice that clear differences in the landscape of the calculated energy levels are not visible in the range of experimentally explored Raman detunings δ R (−0.05 < δ R /Γ < 0.05) of [fig_ref] Figure 5: Raman detuning [/fig_ref]. Indeed in the calculations the values of Raman detuning result irrelevant if much smaller than the light shifts. Thus, in order to observe a difference in the level structure in the calculations we actually need to vary δ R much more than in the experiment (see [fig_ref] Figure 6: Energy of low-scattering states as a function of position [/fig_ref]. The discrepancy could signal that we are performing the calculations with Rabi frequencies larger than the effective experimental values. As a matter of fact the one-dimensional calculation performed considering lin ⊥ lin configuration cannot exactly account for the experimental configuration, which actually consists in three couple of counter-propagating beams, each couple made of circularly polarized σ + /σ − beams. Nonetheless, as already demonstrated 11 the simplified scheme allows to capture the essential mechanism undergoing.
# Conclusions
In summary we have shown that efficient dark-state cooling can be achieved even on the D 2 transition for Rb atoms, thanks to the relatively large hyperfine separations of the upper level which make the F = 2 → F′ = 2 open transition sufficiently isolated from the closed F = 2 → F′ = 3. We reported a thorough experimental characterisation of the grey molasses operation as a function of different experimental parameters such as the intensity of the repumper light, the time duration, and the frequencies of both cooler and repumper. Furthermore, we have pointed out the fundamental role of the phase coherence between the cooler and the repumper laser fields. We find some interesting differences with respect to grey molasses on the D 1 transition, for example in the typical Fano profile shown in [fig_ref] Figure 5: Raman detuning [/fig_ref] where the high-temperature peaks arising for positive Raman detuning reduces at large detunings. Although the calculated energy levels, scattering rates and diabatic couplings provide useful hints, we lack a full explanation for this effect.
Our findings have practical consequences for experiments in cold atoms as they show that the PSD can be increased with Λ-enhanced grey molasses without the drawback of an additional laser source on the D 1 transition. We find phase space densities comparable to the ones achieved with other known techniques as Dark-Spot MOTs [bib_ref] High densities of cold atoms in a dark spontaneous-force optical trap, Ketterle [/bib_ref] [bib_ref] Cold atoms densities of more than 10 12 cm −3 in a..., Radwell [/bib_ref] or Compressed-MOTs [bib_ref] Behavior of atoms in a compressed magneto-optical trap, Petrich [/bib_ref] [bib_ref] Two-color magneto-optical trap for metastable helium, Tychkov [/bib_ref]. In addition, grey molasses achieve efficient optical pumping in F = 1 level, which can be convenient for further experiments. In particular, for mixtures experiments where one species requires grey-molasses cooling, also on the other one it is useful to have it.
# Methods
Experimental procedure: MOT and molasses. We start loading a 3D-MOT from an atomic beam similarly to that described in [bib_ref] Intense slow beams of bosonic potassium isotopes, Catani [/bib_ref]. For the MOT loading, the repumper and cooler light are provided by two distinct diode lasers, overlapped on a single-mode polarisation-maintaining fiber, and they have detunings Δ − . Γ 2 5 [bib_ref] Efficient all-optical production of large 6 Li quantum gases using D 1..., Burchianti [/bib_ref] and Δ − . Γ 0 5 R with respect to the F = 2 → F′ = 3 and F = 1 → F′ = 2 transitions, respectively. The repumper intensity is about 5% of the total 3D-MOT light. A pair of coaxial coils in anti-Helmholtz configuration generates the magnetic field gradient, approximately 14 G/cm in the vertical direction and 7 G/cm along any direction in the horizontal plane. The 3D-MOT consists of six independent laser beams -two counter-propagating beams for each spatial direction -with an intensity of about 6I sat for each beam. We load the 3D-MOT for 5 to 7 seconds, up to a fixed number of 3 × 10 8 atoms. In order to have the same number of atoms collected in the MOT for each experimental run, we collect part of the fluorescence light emitted by the trapped cloud with a photodiode and we stabilise the fluorescence signal to a reference value by modulating, via the intensity of the 2D-MOT beams, the flux of cold atoms loading into the 3D-MOT. Under the assumption that the fluorescence is proportional to the number of atoms, the latter is also stabilised from run to run.
Once the atoms are loaded in the 3D-MOT, we suddenly switch off the magnetic field gradient and change the cooler frequency to the value needed for the grey molasses. Due to technical limitations, this frequency shift cannot be accomplished instantaneously: we linearly ramp the cooler frequency in 2 ms. At the beginning of the ramp, we switch off the (incoherent) repumper light used for the MOT loading and switch on the (coherent) repumper generated by the EOM sideband for the grey molasses.
As for bright molasses, also for grey molasses the lowest temperatures are reached only after proper cancellation of stray magnetic fields, and we estimate to reach residual fields below 0.1 G. For this purpose we use three independent pairs of compensation coils placed along the three 3D-MOT axes.
Finally, in we report the experimental time sequence, as obtained from the parameters optimization described above. In particular, we have found convenient to start the grey molasses with large optical intensitycrucial to capture most part of atoms from the MOT -and then to slowly decrease the beams power with a linear ramp, down to a value lower than the level employed for the MOT loading.
Fitting procedure: time-of-flight curves. After the molasses, all the beams are extinguished and the atoms freely expand; then, after a few ms of TOF, we switch on the MOT beams again and collect the atoms fluorescence on a CCD camera. More precisely, the image is recorded acquiring the fluorescence of the cloud illuminated by the cooler light, thus providing the number of atoms in F = 2 hyperfine state of the ground state level. To measure the total number of atoms, the imaging is forerun by a 200 μs-long pulse of repumper light, which pumps . Experimental sequence. Time dependence of the quadrupole magnetic field (blue), the compensation magnetic field bias (green), the total optical power of the laser beams (red) and the cooler detuning (yellow). We load the MOT for t MOT , then we switch off abruptly the quadrupole magnetic field and ramp the laser frequency and the compensation fields to the molasses value in t ramp = 2 ms, and perform grey molasses for t mol . Then, we suddenly switch off the lights and let the atoms free to expand for t TOF before acquiring a fluorescence image of the atoms in F = 1. To do that, before the image we pump the atoms from the F = 1 to the F = 2 hyperfine state with a repumper pulse of t R = 200 μs, and then switch on the cooler light at the same detuning as the MOT phase -we ramp the frequency back with a second 2 ms ramp when the lights are off during the free expansionand acquire the image.
SciENtiFic RepoRtS | (2018) 8:1301 | DOI:10.1038/s41598-018-19814-z also the atoms in F = 1 to the F = 2 state. Then, we extract the number of atoms in F = 1 state by subtracting from the total number of atoms those in F = 2.
From the recorded images, we measure the size of the cloud σ i (i = x, z) and the number of atoms N.
The dependence of the sizes on the free-fall time duration t TOF is given by the following equation: where the index i = x, z identifies the two spatial directions accessible to fluorescence imaging and σ i 0 are the sizes of the cloud immediately after the grey molasses. Therefore, by fitting the fluorescence images of the expanded atomic cloud we determine σ i for different values of t TOF , and then we extract both σ i 0 and T i . The temperatures extracted are the same within the error; the temperature T reported in this paper is the average between the values obtained in the two directions.
[formula] σ σ = + t k T m t ( ) ( ) (2) i i B i [/formula]
Data availability. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
[fig] Figure 1: Scheme of the energy levels of the D 2 transition in 87 Rb. Red arrows represent the two frequencies employed for the grey molasses; cooler(repumper) is blue-detuned with respect to the F = 2(1) → F′ = 2 transition, with detuning Δ 22 (Δ R ). E hfs denotes the ground-state hyperfine splitting, equal to h × 6834.683 MHz. All other hyperfine splittings are indicated with the corresponding frequencies. SciENtiFic RepoRtS | (2018) 8:1301 | DOI:10.1038/s41598-018-19814-z [/fig]
[fig] Figure 2: Effect of repumper intensity. Left: Temperature T (purple circles) and ratio between atom number after and before the molasses N/N MOT (green triangles) as a function of the intensity ratio I R /I C between repumper and cooler light; Right: PSD measured after grey molasses normalized to the value of the bright molasses (PSD B ). SciENtiFic RepoRtS | (2018) 8:1301 | DOI:10.1038/s41598-018-19814-z at (2π) × 6834.6 MHz. The detunings Δ 22 and Δ R therefore take almost the same value; for simplicity the data in the plots are reported as a function of Δ 22 . [/fig]
[fig] Figure 3: Optimization of molasses duration. Left: Temperature T (purple circles), fraction of remaining atoms N/N MOT (green triangles) and Right: normalized phase space density PSD/PSD B as a function of the duration Δt of the grey molasses. [/fig]
[fig] Figure 4: Influence of detuning: incoherent and coherent repumper. Left: Temperature T (PSD/PSD B is shown in the inset) as a function of the cooler and rempumper detuning. We report both the case of coherent (empty points) and incoherent repumper (filled points). Right: Fractional number of remaining atoms N/N MOT in F = 1 (pink circles) and F = 2 (green triangles); empty (filled) points refer to the case of coherent (incoherent) repumper. SciENtiFic RepoRtS | (2018) 8:1301 | DOI:10.1038/s41598-018-19814-z Raman detuning δ R ≡ Δ RC − E hfs /ħ as the detuning of the repumper light with respect to the Raman condition of the Λ-configuration. The experimental data in Fig. 5 are reported as a function of this quantity. [/fig]
[fig] Figure 5: Raman detuning. Left: Temperature T and Right: normalized phase space density PSD/PSD B (N/N MOT is shown in the inset) as a function of the Raman detuning δ R . In both graphs, we show three different datasets: Δ 22 = 5 Γ (pink circles), Δ 22 = 8 Γ (blue squares) and Δ 22 = 12 Γ (green triangles). [/fig]
[fig] Figure 6: Energy of low-scattering states as a function of position. Purple(green) lines are states with dominant weight in F = 1(2) level; for each state considered, the values plotted in graph are the energy shifts with respect to the corresponding level energy in the absence of the light. The line thickness is proportional to positiondependent scattering rate γ′ defined in text. Left: Energies calculated for δ R = −0.1 Γ; Right: energies calculated for δ R = +0.1 Γ. SciENtiFic RepoRtS | (2018) 8:1301 | DOI:10.1038/s41598-018-19814-z [/fig]
[table] Table 1: Minimum temperature T, fraction of remaining atoms N/N MOT and normalized phase space density PSD/PSD B measured for δ R = −0.01 Γ. [/table]
|
Computational fluid dynamics of the right atrium: Assessment of modelling criteria for the evaluation of dialysis catheters
Central venous catheters are widely used in haemodialysis therapy, having to respect design requirements for appropriate performance. These are placed within the right atrium (RA); however, there is no prior computational study assessing different catheter designs while mimicking their native environment. Here, a computational fluid dynamics model of the RA, based on realistic geometry and transient physiological boundary conditions, was developed and validated. Symmetric, split and step catheter designs were virtually placed in the RA and their performance was evaluated by: assessing their interaction with the RA haemodynamic environment through prediction of flow vorticity and wall shear stress (WSS) magnitudes (1); and quantifying recirculation and tip shear stress (2). Haemodynamic predictions from our RA model showed good agreement with the literature. Catheter placement in the RA increased average vorticity, which could indicate alterations of normal blood flow, and altered WSS magnitudes and distribution, which could indicate changes in tissue mechanical properties. All designs had recirculation and elevated shear stress values, which can induce platelet activation and subsequently thrombosis. The symmetric design, however, had the lowest associated values (best performance), while step design catheters working in reverse mode were associated with worsened performance. Different tip placements also impacted on catheter performance. Our findings suggest that using a realistically anatomical RA model to study catheter performance and interaction with the haemodynamic environment is crucial, and that care needs to be given to correct tip placement within the RA for improved recirculation percentages and diminished shear stress values.
# Introduction
Haemodialysis is used clinically during kidney failure to support blood filtering. This process is enabled by the use of dialysis catheters, devices with a tip placed within the proximal third of the superior vena cava (SVC), the right atrium (RA), or the inferior vena cava (IVC) com/library/the-human-heart-1). The model was truncated at the superior vena cava (SVC), inferior vena cava (IVC) (inlets) and tricuspid valve (TV) (outlet), The geometry was re-scaled to match IVC and SVC physiological mean literature diameters using ANSYS SpaceClaim v.18.2 (Ansys Inc., Canonsburg, PA, USA). Final diameters for the IVC and SVC yielded [bib_ref] A coupled mitral valve-left ventricle model with fluidstructure interaction, Gao [/bib_ref].89 mm and 17.06 mm, respectively, and the area of the TV was 9.95 cm 2 , within in vivo ranges [bib_ref] Assessment of inferior vena cava diameter by echocardiography in normal Indian population:..., Patil [/bib_ref] [bib_ref] Comprehensive Imaging Review of the Superior Vena Cava, Sonavane [/bib_ref] [bib_ref] Geometric determinants of functional tricuspid regurgitation: insights from 3-dimensional echocardiography, Ton-Nu [/bib_ref] [bib_ref] All you need to know about the tricuspid valve: Tricuspid valve imaging..., Huttin [/bib_ref]. Face and edge repair tools from ANSYS SpaceClaim were then used to repair the RA model surfaces, yielding the geometry presented in It is noted that there is an elevated variability in the data reported in literature for the dimensions of the RA, which may be due to variability in subject populations, image acquisition methods and subsequent empirical determination of right atrial volumes [bib_ref] Reference right atrial dimensions and volume estimation by steady state free precession..., Maceira [/bib_ref]. Nonetheless, the volume of our RA geometry, excluding the cava veins, was 175.27 mL, slightly higher than values reported for healthy subjects (upper limit of normality: 170.4 mL [bib_ref] Reference right atrial function determined by steady-state free precession cardiovascular magnetic resonance, Sievers [/bib_ref] , but still within ranges presented by previous clinical papers [bib_ref] Reference right atrial dimensions and volume estimation by steady state free precession..., Maceira [/bib_ref] [bib_ref] Reference right atrial function determined by steady-state free precession cardiovascular magnetic resonance, Sievers [/bib_ref]. Four catheter designs were chosen: catheters A and B have a step tip with and without sideholes, respectively. Catheter C has a split tip and catheter D has a symmetric tip, without side holes (see [fig_ref] Table 1: Catheter dimensions [/fig_ref] for dimensions). The geometries of the four catheters are shown in [fig_ref] Fig 4: RA computational domain with example catheter inserted [/fig_ref]. A total of 8 computational models (including the RA model) were designed, as outlined in 3D geometry files corresponding to catheters designs A, B, C and D were placed in the RA geometry. Catheters were placed through the SVC using ANSYS SpaceClaim, with their entire functional part inside the RA and their venous tip placed well past the SVC in the central region of the RA, to mimic clinical guidelines [bib_ref] A Proposed Simple and Accurate Technique for Optimal Long-Term Hemodialysis Catheter Tip..., Tawk [/bib_ref]. For each catheter model, the structure was removed and only a unified volume including the fluid within the RA and the fluid within the catheter structure was used for subsequent CFD simulations. An example of an inserted catheter into the RA can be seen in All geometries were meshed using ANSYS (Ansys Inc., Canonsburg, PA, USA) and an example of a catheter model mesh within the RA is provided in [fig_ref] Fig 5: Finite-element mesh cross-section of catheter B inserted in RA [/fig_ref] Tetrahedral elements were employed, with a patch conforming method scheme. This scheme allowed the choice of a finer mesh for catheter fluid boundaries in order to achieve greater mesh refinement within this volume. A structured hexahedral mesh was used for the creation of 5 boundary layers in the fluid near the right atrial wall. Using these settings, an average spatial resolution of 0.1 mm was achieved for the solid mesh inside each catheter while 1 mm was achieved for the remaining solid mesh, respectively. Mesh quality was assessed through element skewness and orthogonal quality [fig_ref] Table 2: Mesh settings and quality assessment [/fig_ref]. According to quality criteria, our meshes had excellent skewness features (between 0 and 0.25) and very good orthogonal quality (between 0.70 and 0.95).
A mesh convergence analysis was performed using the RA model, with mesh refinement being achieved by progressively decreasing tetrahedral element size in the ANSYS Meshing Module. The mesh convergence test was performed using ANSYS Fluent 18.2 by running one cardiac cycle. For all meshes, the instantaneous velocity magnitude at a probe located at the centre of the RA (S5 File) and the surface averaged WSS were obtained. The relative error between values obtained under increasing mesh density and the solution obtained when using the finest mesh was then evaluated (S6 File and [fig_ref] Fig 6: Mesh convergence study for RA model [/fig_ref]. Given the complex flow patterns developing inside the RA, we assumed an error below 0.5% as acceptable for the average WSS and below 3% for the probe velocity. Both criteria were met once the model had above 1 million elements.
## Flow governing equations and material properties
Given the importance of using non-Newtonian models for the study of local haemodynamics [bib_ref] Assessment of surface roughness and blood rheology on local coronary haemodynamics: a..., Owen [/bib_ref] , and similarly to the recent paper by [bib_ref] Impact of side-hole geometry on the performance of hemodialysis catheter tips: A..., Owen [/bib_ref] , blood was considered a Non-Newtonian fluid [bib_ref] Modeling Blood Flow Through Intracranial Aneurysms: A Comparison of Newtonian and Non-Newtonian..., Carty [/bib_ref] and modelled using the Bird-Carreau model:
[formula] m ¼ m 1 þ ðm 0 À m 1 Þ½1 þ ðl _ gÞ 2 � ðnÀ 1Þ=2ð1Þ [/formula]
where μ is the blood viscosity, μ 1 is the high shear viscosity, μ 0 is the low shear viscosity, λ is the time constant, _ g is the shear rate and n is the Power law index [bib_ref] Effects of the non-Newtonian viscosity of blood on flows in a diseased..., Cho [/bib_ref]. As per previous studies, the following values were used: λ = 3.313 s, n = 0.3568, μ 0 = 0.056 Pa s and μ 1 = 0.00345 Pa s [bib_ref] Effects of the non-Newtonian viscosity of blood on flows in a diseased..., Cho [/bib_ref] [bib_ref] Non-Newtonian blood flow in human right coronary arteries: steady state simulations, Johnston [/bib_ref] and a blood density of 1060 kg/m 3 . A comparison between this non-Newtonian model
[formula] Re ¼ ρUD mð2Þ [/formula]
where ρ is the blood density, U and μ are the velocity magnitude and viscosity at each boundary, and D is the diameter of the corresponding boundary. Averaged Reynolds numbers have been determined based on time and space-averaged values of U and μ over each boundary, while the minimum and maximum Reynolds values were computed for a time-varying Reynolds number obtained using the averaged U and μ over each boundary [fig_ref] Table 3: Reynolds number study performed for the RA model [/fig_ref]. Since Re << 2300, flow was assumed to be laminar [bib_ref] Direct numerical simulation of transitional flow in a patient-specific intracranial aneurysm, Valen-Sendstad [/bib_ref].
## Boundary conditions
RA model. To accurately represent the pulsatile behaviour of blood flow, a time-dependent physiological pressure two-dimensional waveform was applied at the SVC and IVC inlets [fig_ref] Fig 7: Time-dependent pressure, with diastolic and systolic periods represented, imposed at the inlets [/fig_ref] [36], modelled as a spatially uniform profile. According to this waveform, the complete cardiac cycle corresponds to a period of 0.8 s, and the diastolic and systolic periods last for 0.5 s and 0.3 s, respectively. The TV outlet was set to a constant gauge pressure of 0 Pa. The right atrial wall boundaries were assumed rigid and a no-slip condition was employed at the wallblood interface.
Catheter models. Catheter tip designs employed in our study are shown in The venous and arterial lumens correspond to the placement of inlet and outlet boundary conditions, respectively. For all models with catheters inserted into the RA, the boundary conditions applied for the RA model (and represented in were also included. Although the pressure (and therefore the respective flow rate) generated by a dialysis machine oscillates [bib_ref] Catheter performance, Depner [/bib_ref] , blood flow through the catheter inlet (venous lumen) was set to a constant volume flow rate of 400 ml/min (the maximum flow rate used clinically), together with a constant gauge pressure of 248 mmHg [bib_ref] Catheter performance, Depner [/bib_ref]. At each catheter outlet (arterial lumen), different constant gauge pressures were applied to represent a flow rate within the clinical range, as specified in [fig_ref] Table 4: Gauge pressure values applied at catheter outlets and respective flow rates achieved[5] [/fig_ref].
## Computational settings
ANSYS Fluent 18.2 (Ansys Inc., Canonsburg, PA, USA) was used to implement and solve the CFD simulations, with fluid dynamics being solved using the continuity and incompressible Navier-Stokes equations [bib_ref] Physiology and pathology of the cardiovascular system: a physical perspective, Parker [/bib_ref] under transient conditions. While a single-phase model was assumed for the RA model alone, a multiphase model was set up for all catheter geometries. This choice was made to allow for the quantification of recirculating flow. Two phases simulating blood with identical material properties were defined, where the initial volume of flow entering the RA through the catheter was assumed as one phase (recirculation phase-filtered blood), and the remaining flow was assumed as the other (primary phase-unfiltered blood). Through this process, the flow passing through the venous lumen and the volume of recirculation phase flow entering the arterial lumen of the catheter were monitored and quantifiable. Further details of simulation set-up are summarized in [fig_ref] Table 5: CFD set-up. [/fig_ref].
To determine how many cycles were necessary to achieve temporal convergence, the RA model was run for 8 cycles. The time-averaged mean velocity and pressure were analysed for each cardiac cycle. The relative error (E) between the approximate velocity and pressure solutions for each cycle and the solution from the last cycle (7 th cycle) was then measured, using Eq 3:
[formula] E ¼ v B À v L v L � � � � � � � � � 100%;ð3Þ [/formula]
where v B and v L are the calculated variables for each time period before the last and for the last time period, respectively. Eq 3 was employed for volume-averaged derived velocity and pressure and for probe related quantities (S9 File), with the probe placed at the centre of the RA. A decrease in the error was observed with increasing number of cycles for volume-averaged and probe velocity and pressure : here, the last data point is at the 7 th cycle and it is noted that volume-averaged results have an associated relative error smaller than those of localised velocities and pressure, which are less prone to stabilize. The 4 th cardiac cycle was chosen for result retrieval and quantifications, corresponding to relative errors of 1.28% and 4.12% for volume-averaged velocity and pressure quantities, respectively.
All simulations were run for a total of 3200 time steps, corresponding to 3.2 s (4 cardiac cycles), with the results being retrieved from the last cycle. All numerical simulations were performed using super-computing facilities (BlueBear High Performance Computing, University of Birmingham), with 100 cores for each simulation and 5 GB RAM per core. The solution time for the simulation of the RA model was~10 hours, and~25 hours for the models of the RA with a catheter included.
## Haemodynamics
All simulation results were accessed using Ansys CFD-post. Global right atrial haemodynamics were evaluated in terms of velocity magnitude, flow streamlines, vorticity, pressure and wall shear stress (WSS, both time-averaged and non-time-averaged). Vorticity (ω) is defined as the curl of the velocity field [bib_ref] Transient large strain contact modelling: A comparison of contact techniques for simultaneous..., Espino [/bib_ref] , and describes the local spinning motion of blood flow, as follows
[formula] ω ¼ r � uð4Þ [/formula]
where u is the velocity vector. WSS was calculated and averaged over the whole RA wall. Timeaveraged quantities were also derived considering the cardiac cycle period (0.8 s). All RA results were then validated against the available literature data before using it as a haemodynamic model for catheter performance studies (section 3.1). Catheter performance was evaluated using measures of vorticity, WSS, recirculation and blood shear stress. To calculate recirculation of blood through catheters, the volume fraction (0-1) of the recirculation phase (; r ) is considered: for multiphase catheter models, any measure in a mesh element is weighted between the primary and the recirculation phases. The time-averaged volume fraction of the recirculation phase at the catheter outlet is then defined as where T is the length of the cardiac cycle. For a k th facet of the catheter outlet boundary, the magnitude of the recirculation fraction can then be defined as a mass-weighted average of ; k through the catheter outlet,
[formula] ; r ¼ 1 T Z T 0 ; r dtð5ÞR f ¼ 1 _ m T X n k¼1 ; k _ m kð6Þ [/formula]
where _ m T and _ m k are the total mass flow rate over the catheter outlet boundary and the mass flow rate for a k th facet, respectively. The latter is defined by
[formula] _ m k ¼ ρðu k � A k Þð7Þ [/formula]
where u k and A k are the k th facet velocity and area vectors, respectively. Previous studies have noted higher levels of platelet activation at the venous lumen tip in comparison with the arterial one, with the latter yielding small differences amongst catheter designs [bib_ref] Comparison of Symmetric Hemodialysis Catheters Using Computational Fluid Dynamics, Clark [/bib_ref]. To better observe any marked differences between models, we chose the venous tip for analysis and quantification of shear stress, which can be defined as
[formula] τ ¼ μ � j� ij j;ð8Þ [/formula]
where τ is the shear stress and j� ij j is the magnitude of the strain rate. The strain rate tensor and its magnitude are defined by Eqs (9) and (10), respectively,
[formula] � ij ¼ 1 2 @u j @x i þ @u i @x j ! ;ð9Þj� ij j ¼ ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi 2� ij � ij ; qð10Þ [/formula]
where u i and u j are the velocity vectors in the i and j directions and x i and x j are the spatial coordinates in the i and j directions. Furthermore, rectangular prism volumes were created at each venous tip [bib_ref] Particle image velocimetry-validated, computational fluid dynamics-based design to reduce shear stress and..., Mareels [/bib_ref] by defining (x, y, z) boundaries in ANSYS CFD-post (example of volume definition in [fig_ref] Fig 9: Example of volume definition to be placed at the tip [/fig_ref]. The size of all tip volumes is specified in [fig_ref] Table 6: Dimensions for the creation of tip volumes for all catheters [/fig_ref]. For catheter A, the tip volume included all orifices of the tip.
At this tip volume, volume-averaged shear stress was calculated, according to: where V is the tip volume. The tip volume (%) where τ � 10 Pa was evaluated to identify the potential for platelet activation [bib_ref] Particle image velocimetry-validated, computational fluid dynamics-based design to reduce shear stress and..., Mareels [/bib_ref] [bib_ref] Effects of exercise training and deconditioning on platelet aggregation induced by alternating..., Wang [/bib_ref].
[formula] τ ¼ 1 V Z tip τ dV;ð11Þ [/formula]
# Results
## Right atrium model validation
Right atrial CFD results were compared against in vivo and in vitro results from the literature; this comparison focused on data for a healthy RA model without the presence of a catheter. This approach to validation meant that more data was available from literature for comparison. Blood velocity magnitudes were within the range of those presented by in vivo reports.
The time-and volume-averaged velocity magnitude was 0.192 m/s, while the minimum and maximum values of the volume-averaged velocity magnitude oscillated between 0.160 m/s and 0.233 m/s, respectively. The average value was in the range of reported values by previous clinical (0.174 ± 0.027 m/s) [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] and in vitrostudies. In addition, the predicted time-and volume-averaged pressure was 1.18 mmHg, while the minimum and maximum values of the volume-averaged pressure varied between 0.55 and 1.88 mmHg, respectively. Maximum pressure values reached 4.55 mmHg. Clinical guidelines specify that, for an IVC diameter < 21 mm (as is the case of the model), the normal time-and volume-average RA pressure is 3 mmHg, with minimum and maximum volume-averaged values varying between 0 and 5 mmHg [bib_ref] Guidelines for the echocardiographic assessment of the right heart in adults: a..., Rudski [/bib_ref]. Therefore, the obtained computational pressure is within estimated clinical values.
The model also predicted characteristic flow patterns within the RA, with the presence of a vortex originated from the IVC flow and SVC flow swirling around it in a helical fashion [fig_ref] Fig 10: Right atrial flow patterns [/fig_ref]. This is also corroborated by clinical studies, which speculate that this swirling motion optimises blood flow within the heart [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref] [bib_ref] Assessment of viscous energy loss and the association with three-dimensional vortex ring..., Elbaz [/bib_ref]. The predicted volume-averaged vorticity for the RA was 44.1 s -1 , which is within the range of in vivo predictions (37-54 s -1 ) [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref]. [fig_ref] Fig 10: Right atrial flow patterns [/fig_ref] shows the presence of both vortical and helical features, with counter-rotating flow filling the RA (positive and negative helical structures). Although a purely clockwise vortex is most common [bib_ref] Asymmetric redirection of flow through the heart, Kilner [/bib_ref] [bib_ref] 4D cardiovascular magnetic resonance velocity mapping of alterations of right heart flow..., Francois [/bib_ref] , a spectrum of right atrial flow patterns exists in the structurally normal heart, including vortical, helical and multiple vortical flow [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] , consistent with the predicted flow patterns.
The time-evolution of flow rates through the SVC, IVC and TV over a cardiac cycle period are presented in [fig_ref] Fig 11: Blood flow rate profiles over one cardiac cycle at SVC, IVC and... [/fig_ref] Similar to the literature, both SVC and IVC have similar flow rate waveforms, with a two-peaked shape (a peak at ventricular systole and a smaller one at ventricular diastole) [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] [bib_ref] Age-related changes of right atrial morphology and inflow pattern assessed using 4D..., Wehrum [/bib_ref]. Literature shows that the range of these flow rate waveforms can greatly vary amongst a population sample and with age [bib_ref] Age-related changes of right atrial morphology and inflow pattern assessed using 4D..., Wehrum [/bib_ref]. Our predictions yield maximum systolic flow rate values of 106 ml/s and 120 ml/s for the IVC and SVC. While the obtained waveform and maximum for the IVC are consistent with clinical predictions (maximum value of 151.1 ± 55.3 ml/s at systole for a group of healthy adults aged 20-39 years), the SVC waveform and maximum value are overestimated (maximum value increased by 22% in comparison with the maximum standard deviation observed in the literature [bib_ref] Age-related changes of right atrial morphology and inflow pattern assessed using 4D..., Wehrum [/bib_ref].
TV function was not included in the model (see Limitations section); hence, the TV flow rate waveform follows a similar pattern in comparison with that of the IVC and SVC, with a peak flow rate of 225 ml/s. However, the current literature does not present many examples of the expected flow rate through the TV making any comparison difficult. One study shows that there is a phase shift in the location of the TV flow rate peaks, which should appear during the ventricular diastolic period [bib_ref] Patients with pulmonary fibrosis: cardiac function assessed with MR imaging, Kroft [/bib_ref]. This did not occur in our flow rate predictions, likely because TV function was not modelled.
The haemodynamic results obtained for the RA, including WSS and vorticity, are further discussed in Section 3.2, as well as compared with those obtained by our CFD catheter models.
## Catheter insertion
Recirculation, vorticity, time-averaged WSS magnitude and shear stress results are presented in [fig_ref] Table 7: Haemodynamic predictions for all catheter models [/fig_ref] for the RA model and all catheter designs.
Impact of catheter insertion on RA haemodynamics. The time evolution of WSS and vorticity in the RA model showed similar trends [fig_ref] Fig 12: Time evolution of spatially averaged WSS and volume-averaged vorticity [/fig_ref]. A correlation is present between WSS and vorticity, quantified based on a Pearson correlation coefficient of 0.87.
Time and volume-averaged vorticity increased in all catheter models in comparison with the RA model [fig_ref] Table 7: Haemodynamic predictions for all catheter models [/fig_ref] and [fig_ref] Fig 3: Diagrammatic overview of all computational models developed [/fig_ref]. Catheter designs yielded similar vorticity values (54-57.20 s -1 ), with a 29.71% maximum increase from RA vorticity. Catheter C had the greatest average flow vorticity, as indicated by increased vorticity values during early systole and late diastole in comparison with other designs [fig_ref] Fig 3: Diagrammatic overview of all computational models developed [/fig_ref]. Changing the catheter tip placement or rotating it did not greatly impact the average RA vorticity for designs A and D [fig_ref] Fig 3: Diagrammatic overview of all computational models developed [/fig_ref] , yielding absolute average differences of 2.22% and 1.85%, respectively.
As observed in [fig_ref] Table 7: Haemodynamic predictions for all catheter models [/fig_ref] , time-averaged WSS on the RA wall was not markedly affected for catheters B and C. Catheters A1, A2 and A3 suffered time-averaged WSS percentage increases ranging between 6 and 9%, while catheters D1 and D2 had the greatest percentage increases (13-14%), in comparison with the value obtained for the RA model.
Time-averaged WSS magnitude distributions are presented on [fig_ref] Fig 4: RA computational domain with example catheter inserted [/fig_ref] Areas with elevated WSS are located around the SVC and IVC inlets in all models, possibly due to local diameter reductions. Both sites of low and high magnitudes are present on the wall of the RA model, with high magnitude locations possibly corresponding to regions of elevated vorticity and helicity. The central region of the RA has sites of increased WSS magnitude for all models, whose surface distribution changes in the catheter models according to the blood flow jet pattern from each catheter tip. Moreover, designs A and D were associated with different sites of increased WSS, as observed below the SVC junction (circled on [fig_ref] Fig 4: RA computational domain with example catheter inserted [/fig_ref].
Analysis of recirculation. All catheter models yielded flow recirculation [fig_ref] Table 7: Haemodynamic predictions for all catheter models [/fig_ref] : while B was associated with the highest recirculation percentage (43.7%), A and C yielded recirculation > 5%. [fig_ref] Fig 3: Diagrammatic overview of all computational models developed [/fig_ref] further information on the performance of these catheters: the side holes present in A allow for venous lumen flow to enter the RA with different trajectories, which seems to assist in a better mixing of this flow with the chamber flow and prevents it from entering the arterial lumen. The lack of side holes in B, however, seems to enhance the amount of flow returning through the catheter arterial lumen. Design D gave rise to the lowest percentages of recirculating fluid (< 0.30%), meeting the design requirements of less than 1%
## Plos one
giving rise to a greater concentration of venous flow at the base of the RA and near the IVC junction, respectively.
Analysis of shear stress. C and D catheter models were characterized by the lowest volume time-averaged shear stress at the tip, while the A and B designs led to the highest ones [fig_ref] Table 7: Haemodynamic predictions for all catheter models [/fig_ref]. C was associated with a small percentage of volume of τ > 10 Pa (15.70%), while the other designs yielded values in the same range (above 28%). Moreover, and as observed in [fig_ref] Table 7: Haemodynamic predictions for all catheter models [/fig_ref] and [fig_ref] Fig 6: Mesh convergence study for RA model [/fig_ref] changing tip placement affected the predicted shear stress for A and D. A tip placement closer to the atrium wall (Position 2) increased the volume-averaged shear stress and percentage of volume of τ > 10 Pa for design A by 20.15% and 18.02%, respectively, but decreased these quantities for design D by 12.07% and 0.70%, respectively. Rotating catheter A (Position 3) greatly improved these outcomes, but there was still an increase of 6.20% in the volume-averaged shear stress and of 1.06% on the percentage of volume of τ > 10 Pa in comparison with the first placement. The shear stress profile from [fig_ref] Fig 6: Mesh convergence study for RA model [/fig_ref] temporal trends, and interestingly, for design D, the first position is associated with unsteady shear stress through time (also observed on [fig_ref] Fig 6: Mesh convergence study for RA model [/fig_ref] while the second one yields smoother shear stress changes. [fig_ref] Fig 6: Mesh convergence study for RA model [/fig_ref] shows that, while A and B designs yield relatively constant temporal shear stress, for C this stress increased near end-systole and decreased through diastole. Shear stress equal to 10 Pa were localized to both inlet and outlet boundaries, as well as side-holes [fig_ref] Fig 7: Time-dependent pressure, with diastolic and systolic periods represented, imposed at the inlets [/fig_ref]. The venous lumen exit, however, had a greater proportion of shear stress equal to 10 Pa for all catheters.
# Discussion
## Main study findings
This is the first study to develop and validate a CFD model of the RA to assess catheter design performance. Key haemodynamics have been quantified for four distinct catheter designs. The obtained results suggest the following findings:
- The haemodynamic predictions for the RA model are consistent with in vivo data available in literature [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref] [bib_ref] 4D cardiovascular magnetic resonance velocity mapping of alterations of right heart flow..., Francois [/bib_ref] and with clinical guidelines for right heart assessment [bib_ref] Guidelines for the echocardiographic assessment of the right heart in adults: a..., Rudski [/bib_ref] , therefore, verifying our computational model;
- Catheter insertion induces increased vorticity and alterations in time-averaged WSS in RA haemodynamics;
- Recirculation is present in all catheter designs, with only the symmetric design D complying with required specifications (< 1%);
- The presence of side holes decreases the amount of recirculating flow in step designs, as given by lower recirculation percentages (6.45-9.32%) in design A when compared to design B (43.7%); - Catheters working in reverse mode (step designs) are associated with reduced performance, assessed through greater recirculation percentages and average shear stress values;
- Elevated tip shear stress Pa) is present in catheter designs, which can induce platelet activation and aggregation and subsequently thrombosis [bib_ref] Platelets and shear stress, Kroll [/bib_ref] ;
- Different catheter tip placements impact on performance, as given by altered recirculation percentages, tip shear stress values and percentage of tip volume with τ > 10 Pa;
- The catheter design with best performance is the symmetric one, associated with low recirculation and shear stress values.
## Computational model validation
Both the function and geometry of the RA are heterogeneous across patients [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref]. This variability has not been evaluated in our study, instead we assumed our RA model to provide a representative model of the RA (with/without dialysis catheters). The advantage of this approach is that it enables a direct comparison of the range of catheters evaluated (including positioning). Similarly to previous heart modelling studies, the outlet was not extended as the location of the valve was assumed as the outflow boundary [bib_ref] Computational Flow Dynamic Analysis of Right and Left Atria in Patent Foramen..., Rigatelli [/bib_ref] [bib_ref] Hemodynamics in the Left Atrium and Its Effect on Ventricular Flow Patterns, Vedula [/bib_ref] , as extending the outlet boundary beyond the location of the valve would give rise to incorrect computational predictions as a valve is found at that position (e.g. alteration of pressure gradients near the valve). Moreover, the SVC and IVC inlets are placed 4 and 2 diameters away from the main chamber, respectively. To assess the independence of computational predictions from the domain size, additional simulations were solved for the RA model, with this having the IVC extended and the IVC inlet placed at 4 diameters away from the main chamber. Volume-averaged velocity and vorticity quantities, as well as area-averaged WSS, were evaluated and average relative errors between predictions with and without IVC extension were 4.77%, 4.52% and 2.09% for velocity, vorticity and WSS, respectively. These relative errors are deemed acceptable and justify the choice of our RA domain for computational simulations.
Although blood flow through the RA was assumed laminar, it is known that, for some blood vessels (cranial artery bifurcation, for example), turbulence occurs at Reynolds numbers as low as 400 [bib_ref] The hemodynamic importance of the geometry of bifurcations in the circle of..., Roach [/bib_ref]. However, maximum calculated Reynolds numbers were lower than this value, and so the laminar assumption was considered valid. An advantage of using a RA CFD model to evaluate catheter performance is the possibility to assess the interaction between flow exiting the venous lumen of the catheter and the atrial wall, which is not feasible with simplified models [bib_ref] Computational fluid dynamics-analysis of the Niagara hemodialysis catheter in a right heart..., Mareels [/bib_ref]. However, in silico and in vitro right atrial flow literature is limited [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref] , which does not enable extensive model validation. In vivo data for the RA, on the other hand, is based on small data sets with inter-individual variability [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref]. Nonetheless, the comparison of our RA model haemodynamic predictions with in vivo measurements was considered fundamental and in vitro/in silico results were second choices for validation, due to the fact that any obtained data is subject to individual experimental bias or computational assumptions.
The choice of catheter inlet and outlet flow rates and pressures was based on literature values: clinically realistic flow rates range between 200 and 400 ml/min [bib_ref] Catheter performance, Depner [/bib_ref]. Further insight on catheter performance under varying flow rates can be considered future work. Moreover, on the venous side, blood flow through the catheter is driven by the positive pressure generated by the blood pump, while on the arterial side the driving force is the negative pressure generated by the same pump. This means that pressure values typically range between -250 and +250 mmHg on arterial and venous sides, respectively, justifying our boundary condition choices for catheter models [bib_ref] Catheter performance, Depner [/bib_ref]. On the other hand, the catheter structure was assumed rigid, with our model not accounting for catheter tip movement during the cardiac cycle, which is significant for split tips working in reverse mode [bib_ref] Hemodialysis catheter tip design: observations on fluid flow and recirculation, Vesely [/bib_ref]. A fluid-structure interaction approach for catheter deformation inside the RA haemodynamic environment needs to be employed; however, this type of approach is challenging and its application towards the study of catheter performance remains a current open problem.
## Vorticity and wss in the ra are altered with catheter insertion
Temporal WSS magnitudes were statistically correlated with vorticity in the RA model, which shows that tracking the temporal behaviour of vortex structures may provide complimentary information on the WSS. Similar to previous studies, we can hypothesize that the influence of rinsing motion of increasing vortex during the systolic phase is associated with increasing WSS, thereby avoiding thrombus formation within the RA. Predicted time-averaged WSS values, however, were higher than those previously obtained by a 4D cardiac MRI study [bib_ref] 4D cardiovascular magnetic resonance velocity mapping of alterations of right heart flow..., Francois [/bib_ref]. However, the accuracy of MRI measurements is narrowed by low temporal and spatial resolutions of the order of 40 ms and 2 mm 3 , respectively [bib_ref] Regional variation of wall shear stress in ascending thoracic aortic aneurysms, Rinaudo [/bib_ref]. This causes an averaging of the in vivo measured velocity field, yielding spurious errors in the calculated velocity gradients at the wall-blood interface [bib_ref] Regional variation of wall shear stress in ascending thoracic aortic aneurysms, Rinaudo [/bib_ref] [bib_ref] MRI measurement of time-resolved wall shear stress vectors in a carotid bifurcation..., Papathanasopoulou [/bib_ref]. Therefore, MRI-derived WSS magnitude values are usually underestimated [bib_ref] Regional variation of wall shear stress in ascending thoracic aortic aneurysms, Rinaudo [/bib_ref] , which makes computational modelling advantageous in this matter [bib_ref] MRI measurement of time-resolved wall shear stress vectors in a carotid bifurcation..., Papathanasopoulou [/bib_ref].
Increased vorticity, either time-averaged or through the cardiac cycle, was found in all catheter models with respect to the RA model. No previous studies on catheter design performance have analysed this aspect; however, efforts have been made to understand the connection between vorticity and cardiac function and efficiency [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref]. As mentioned, the presence of vortices in the healthy RA seems to optimise blood flow and cardiac efficiency within the right heart [bib_ref] 4D flow MRI assessment of right atrial flow patterns in the normal..., Parikh [/bib_ref] [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref]. However, previous work has shown: increased right atrial vorticity in patients after repair of Tetralogy of Fallot, with these possessing additional diastolic vortices that impacted on right ventricular flow [bib_ref] Assessment of intracardiac flow and vorticity in the right heart of patients..., Hirtler [/bib_ref] ; and altered right atrial vorticity in patients with right ventricular diastolic dysfunction [bib_ref] Vorticity is a marker of right ventricular diastolic dysfunction, Fenster [/bib_ref]. [bib_ref] Vorticity is a marker of right ventricular diastolic dysfunction, Fenster [/bib_ref] suggested that, although premature, quantifying vorticity could be useful to: 1) correlate to right heart pathologies, and 2) serve as a non-invasive biomarker for the assessment of both haemodynamic and bioenergetics response to therapy [bib_ref] Vorticity is a marker of right ventricular diastolic dysfunction, Fenster [/bib_ref]. Based on our results and the outcomes of previous studies, we speculate that catheter insertion in the RA may alter normal blood flow within this chamber [bib_ref] 4D cardiovascular magnetic resonance velocity mapping of alterations of right heart flow..., Francois [/bib_ref]. However, objective clinical outcomes derived from this speculative hypothesis would have to be further studied.
Previous clinical [bib_ref] Bicuspid aortic cusp fusion morphology alters aortic three-dimensional outflow patterns, wall shear..., Mahadevia [/bib_ref] and computational [bib_ref] Bicuspid aortic valve aortopathies: An hemodynamics characterization in dilated aortas, Oliveira [/bib_ref] [bib_ref] Computational comparison of regional stress and deformation characteristics in tricuspid and bicuspid..., Cao [/bib_ref] studies have linked abnormal flow with high WSS, as well as different WSS distributions, in the ascending aorta. Abnormally high WSS has, in fact, been assumed as a trigger for aortic dilation in congenital diseases by changing the wall tissue mechanical properties [bib_ref] Ascending aortic dilatation associated with bicuspid aortic valve: pathophysiology, molecular biology, and..., Tadros [/bib_ref] [bib_ref] Aortic dilation in bicuspid aortic valve disease: flow pattern is a major..., Bissell [/bib_ref] [bib_ref] Bicuspid aortic valve is associated with altered wall shear stress in the..., Barker [/bib_ref]. Our results show a change in time-averaged WSS, as well as magnitude distributions in the atrial wall surface, with catheter insertion in the RA. Indeed, the symmetric tip (design D) was associated with the highest increase. Moreover, step and symmetric tips (designs A and D) gave rise to the creation of different high WSS sites. Given the previous findings mentioned above, we can hypothesize that these WSS alterations could possibly be associated with right atrial enlargement onset and progression at specific sites.
## Catheter performance
Predicted time-averaged recirculation values for all models are within the same range as those obtained by previous studies [bib_ref] Hemodialysis catheter tip design: observations on fluid flow and recirculation, Vesely [/bib_ref] [bib_ref] Computational flow dynamics and preclinical assessment of a novel hemodialysis catheter, Clark [/bib_ref] [bib_ref] Comparison of Symmetric Hemodialysis Catheters Using Computational Fluid Dynamics, Clark [/bib_ref]. Similarly to a previous study, the symmetric design (D) yielded negligible (< 0.5%) recirculation [bib_ref] Comparison of Symmetric Hemodialysis Catheters Using Computational Fluid Dynamics, Clark [/bib_ref] , and the lowest of all models, which is associated with a better separation of filtered and unfiltered blood. Moreover, designs working in reverse mode (step tips) gave rise to higher recirculation percentages, in comparison with symmetric and split designs. In fact, the literature shows the presence of up to 86% of recirculation for catheters working in reverse configuration [bib_ref] Hemodialysis catheter tip design: observations on fluid flow and recirculation, Vesely [/bib_ref] [bib_ref] Optimizing dialysis delivery in tunneled dialysis catheters, Pannu [/bib_ref] , which can validate our highest recirculation value (43.7%) for step design B. Moreover, and similarly to previous work [bib_ref] Hemodialysis catheter tip design: observations on fluid flow and recirculation, Vesely [/bib_ref] [bib_ref] Comparison of Symmetric Hemodialysis Catheters Using Computational Fluid Dynamics, Clark [/bib_ref] , the presence of side holes in step tip design (A and B) impacted in recirculation rates by diminishing time-averaged recirculating flow. We hypothesize that the presence of side holes in a step tip may improve performance by allowing flow deflection from the distal tip, as suggested elsewhere [bib_ref] Comparison of Symmetric Hemodialysis Catheters Using Computational Fluid Dynamics, Clark [/bib_ref].
Shear stress characteristics were also evaluated to study the tendency of each catheter design to cause shear-induced platelet activation and aggregation. Platelet activation has been shown to induce device thrombogenicity and they experience shear-induced activation at a larger rate than what is required for haemolysis of red blood cells [bib_ref] In vitro blood damage by high shear flow: human versus porcine blood, Klaus [/bib_ref]. Our predicted shear stresses were in the same order of magnitude as presented elsewhere, although percentages of tip volume above 10 Pa were much lower [bib_ref] Particle image velocimetry-validated, computational fluid dynamics-based design to reduce shear stress and..., Mareels [/bib_ref]. This could, however, be due to the variability in tip volume definition. Nonetheless, platelet activation has been observed with 0.1 � τ � 20 Pa [bib_ref] Platelets and shear stress, Kroll [/bib_ref]. Here, we used a middle value as a threshold (10 Pa), not being overly conservative, as it is a relative measure which is being used ultimately to compare catheters.
All catheter models yielded shear stress above 10 Pa, and such stress was mainly observed at the venous tip and side holes in the step design. The elevated shear stress location was similar to that observed in a previous in vivo study, which showed the formation of a fibrin plaque on catheter surface around a venous side hole [bib_ref] Blood flow in hemodialysis catheters: a numerical simulation and microscopic analysis of..., Lucas [/bib_ref]. This suggests that all designs have a potential for shear-induced platelet activation and subsequently thrombosis.
Catheters working in reverse mode (step tip designs) yielded the highest recirculation and time-averaged shear stress values. According to this, the highest potential for shear-induced platelet activation and worse recirculation outcomes were observed for step designs in reverse mode, implying that, in a clinical scenario, the use of standard mode should be targeted. The symmetric tip, however, was shown to have the best performance, as given by the lowest recirculation and shear stress values. Different tip placements also yielded different recirculation percentages, with a tip placement closer to the RA wall giving rise to greater recirculating flow. This is the first computational study which considers how different catheter tip positions impact on the respective performance, as well as how they affect flow within the RA. However, the optimal positioning of an haemodialysis catheter is a continuous subject of debate, with elevated clinical variability [bib_ref] Central venous catheter tip position: a continuing controversy, Vesely [/bib_ref] and changing medical guidelines [bib_ref] A Proposed Simple and Accurate Technique for Optimal Long-Term Hemodialysis Catheter Tip..., Tawk [/bib_ref]. Previous studies noted that this positioning is crucial to prevent/diminish recirculation [bib_ref] Vascular access for hemodialysis: current perspectives, Santoro [/bib_ref] and it is known that such percentage impacts on the efficiency of the haemodialysis treatment [bib_ref] Measuring central venous structures in humans: implications for centralvein catheter dimensions, Twardowski [/bib_ref]. Per our results, a tip placement at the mid-level of the RA with its arterial lumen facing the mediastinum yields lower (but still significant) recirculation percentages, which is in agreement with the latest medical guidelines [bib_ref] KDOQI clinical practice guidelines and clinical practice recommendations-2006 updates, Gilmore [/bib_ref]. Given this, care should be taken with catheter tip placement and orientation within the RA.
# Limitations
There are several limitations in this study, with most arising from the rigid walls and geometry of the RA reconstructed from dimensions in literature. Whilst such reconstruction lacks physiological accuracy, it presents a superior platform for quantitative comparisons of catheter performance compared to previously cylindrical models. Moreover, the RA model was used to mimic a healthy case, which may not be accurate for dialysis patients, especially concerning measures as central pressure [bib_ref] Central venous catheters for haemodialysis: looking for optimal blood flow, Jean [/bib_ref]. Nonetheless, the use of such a model for the study of multiple catheter designs enables direct and objective comparison of their performance, including the evaluation of blood mixing (between blood filtered through dialysis and blood which was not filtered), not feasible through a cylinder model.
The absence of atrial contraction and TV opening/closing during the cardiac cycle was a major limitation, since the movement of the RA walls affects the blood flow patterns and mixing within the chamber, affecting catheter implementation. By employing rigid RA walls, and having into account the mass conservation balance, all flow entering this chamber must be equal to the flow exiting the TV, which is not physiologically accurate. Moreover, fixed-wall simulations oversimplify inflow-outflow boundary conditions and ignore atrial-ventricular interactions; they have been shown to yield different flow fields and stasis mapsand overestimate instantaneous quantities (e.g. flow velocity; WSS) when compared to moving wall simulations [bib_ref] Influence of arterial mechanical properties on carotid blood flow: Comparison of CFD..., Lopes [/bib_ref] ; however, time-averaged quantities have been shown not to greatly vary [bib_ref] Fluid-structure interaction analysis of a patient-specific right coronary artery with physiological velocity..., Torii [/bib_ref]. The use of a fixed RA model may explain the overestimation of the velocity of blood flow through the IVC and SVC, their flow rate and the elevated WSS magnitudes observed in those vessels.
Whilst including RA wall motion may improve flow/pressure predications [bib_ref] Basic mathematical models and motivations, Formaggia [/bib_ref] , very few studies provide insight on RA wall movement [bib_ref] Right atrial function by speckle tracking echocardiography in atrial septal defect: Prediction..., Vitarelli [/bib_ref] [bib_ref] Coupled vs. uncoupled pericardial constraint: effects on cardiac chamber interactions, Takata [/bib_ref] and experimental data on the mechanical properties of the RA wall is currently limited: previous studies mention that in vivo properties for the heart walls can be up to four orders of magnitude different from ex vivo properties [bib_ref] On the effect of prestrain and residual stress in thin biological membranes, Rausch [/bib_ref]. Incorporating a non-rigid wall would, therefore, introduce a range of variables for which data is currently lacking. Given that RA models are still an emerging method, and despite these limitations, fixed-wall simulations can still be useful to assess the essential characteristics of blood flow within the RA.
The SVC blood flow rate waveform and maximum value were overestimated in comparison with the literature. However, there is elevated variability in patient data for blood flow rate through the SVC and IVC: for example, the difference between average and the maximum standard deviation values for the IVC in literature is 33% [bib_ref] Age-related changes of right atrial morphology and inflow pattern assessed using 4D..., Wehrum [/bib_ref]. Moreover, since other hemodynamic features in the RA from our study are consistent with the literature, we accept this as a limitation of the study which does not greatly impact on the remaining computational predictions concerning the evaluation of catheter performance, especially since all catheters have been compared using the same model.
The function of the TV was not modelled. This yielded a different flow rate waveform in comparison with those from the literature [bib_ref] Patients with pulmonary fibrosis: cardiac function assessed with MR imaging, Kroft [/bib_ref] , potentially giving rise to flow patterns and mixing within the RA which differ from a model with valve opening/closing. Moreover, the Neumann assumption may not have been satisfied during the diastolic period, when the TV is open, due to the location of the outlet and valve function not being included; however, it is when the valve is closed during the systole. Including TV function in the model would most likely lead to a more accurate flow rate waveform through the cardiac cycle, as well as increase the pressure drop due to valve closure (which could increase the range of variation of pressure values through one cardiac cycle). Moreover, valve closure would allow the RA to act as a better "reservoir" during systole, leading to a greater flow volume within the RA during this period (when the TV is closed). This could enhance the presence of vortices within the RA (increasing vorticity predictions during systole) and possibly give rise to retrograde flow in the cava veins. Nonetheless, our study focuses on the use of a RA model to evaluate catheter designs and incorporating a fluid-structure interaction valve model would fall outside its scope. In addition, our model enabled good approximations of flow velocity and flow patterns within the RA, in comparison with the literature. Our current RA model has also enabled the study of catheter performance including evaluation of mixing of blood (at the level of the TV), which had, and had not, been filtered via haemodialysis.
# Conclusion
In this study we present a model which provides realistic predictions of haemodynamics in the right atrium, subsequently aiding assessment of haemodialysis catheter performance. Our model shows that the symmetric tip design is associated with the best haemodynamic results, given by its low recirculation and shear stress values, while the step tip designs working in reverse mode gave rise to the worst haemodynamic outcomes. Moreover, the presence of side holes at the tip helped diminish recirculating flow, suggesting that, in the design process of a step tip, this feature should be looked into to improve its performance. In addition, catheter performance was affected by different tip placements, showing that correct positioning should be accounted for when placing the device in the RA.
## Supporting information
[fig] Fig 3: Diagrammatic overview of all computational models developed. Acute refers to temporary catheter placement and chronic to permanent catheter placement, this is of relevance to the clinical use of the catheters.https://doi.org/10.1371/journal.pone.0247438.g003 [/fig]
[fig] Fig 4: RA computational domain with example catheter inserted. https://doi.org/10.1371/journal.pone.0247438.g004 [/fig]
[fig] Fig 5: Finite-element mesh cross-section of catheter B inserted in RA.https://doi.org/10.1371/journal.pone.0247438.g005 [/fig]
[fig] Fig 6: Mesh convergence study for RA model. https://doi.org/10.1371/journal.pone.0247438.g006 [/fig]
[fig] Fig 7: Time-dependent pressure, with diastolic and systolic periods represented, imposed at the inlets (adapted from Cohen et al. 1986). This boundary condition can be generated with the code from the S8 File. https://doi.org/10.1371/journal.pone.0247438.g007 [/fig]
[fig] Fig 9: Example of volume definition to be placed at the tip. https://doi.org/10.1371/journal.pone.0247438.g009 [/fig]
[fig] Fig 10: Right atrial flow patterns: Streamline fields representing velocity magnitude are presented (left), as well as isosurfaces representing vorticity (middle) and helicity (right), at the beginning of systole (t = 0.25 s). https://doi.org/10.1371/journal.pone.0247438.g010 [/fig]
[fig] Fig 11: Blood flow rate profiles over one cardiac cycle at SVC, IVC and TV boundaries (generated with data from the S10 File).https://doi.org/10.1371/journal.pone.0247438.g011 [/fig]
[fig] Fig 12: Time evolution of spatially averaged WSS and volume-averaged vorticity (generated with data from S11 File).https://doi.org/10.1371/journal.pone.0247438.g012 [/fig]
[fig] Fig 15: Time-averaged volume fraction of filtered blood (recirculation phase) within the RA for all catheter models. https://doi.org/10.1371/journal.pone.0247438.g015 [/fig]
[fig] Fig 14: Time-averaged WSS [Pa] for the whole RA domain. https://doi.org/10.1371/journal.pone.0247438.g014 [/fig]
[fig] Fig 16: Volume-averaged shear stress profile through the cardiac cycle for all catheter venous lumen tips (generated with data from S13 File). (a) All designs are present; (b) A and D tip placement changes impact on tip shear stress. https://doi.org/10.1371/journal.pone.0247438.g016 [/fig]
[fig] Fig 17: Isosurface regions where blood shear stress is 10 Pa for all models at the beginning of systole t = 0.25 s. https://doi.org/10.1371/journal.pone.0247438.g017 [/fig]
[fig] S1: File. STL file representing the geometry of catheter A. (STL) S2 File. STL file representing the geometry of catheter B. (STL) S3 File. STL file representing the geometry of catheter C. (STL) S4 File. STL file representing the geometry of catheter D. (STL) S5 File. Probe location for the mesh independence study. (DOCX) S6 File. Values and calculations for the mesh independence study. (DOCX) S7 File. Comparison of Newtonian and non-Newtonian Bird-Carreau models. (DOCX) S8 File. C code that generates the inlet boundary condition pressure waveform. (C) S9 File. Velocity and pressure values used for temporal convergence study. (XLSX) 0 File. Velocity, area, and respective flow rate values used to generate flow rate temporal profiles at SVC, IVC and TV boundaries. (XLSX) 1 File. Spatially averaged WSS and volume-averaged vorticity values for the RA model. (XLSX) 2 File. Volume-averaged vorticity values for all models. (XLSX) 3 File. Volume-averaged shear stress values for all catheter models. (XLSX) Author Contributions Conceptualization: Diana C. de Oliveira, David G. Owen, Shuang Qian, Naomi C. Green, Daniel M. Espino. Data curation: Diana C. de Oliveira. Formal analysis: Diana C. de Oliveira. Funding acquisition: Naomi C. Green, Daniel M. Espino. Investigation: Diana C. de Oliveira. Methodology: Diana C. de Oliveira, David G. Owen, Shuang Qian, Naomi C. Green, Daniel M. Espino. Project administration: Naomi C. Green, Daniel M. Espino. Software: David G. Owen. Supervision: Naomi C. Green, Daniel M. Espino, Duncan E. T. Shepherd.Validation: Diana C. de Oliveira, David G. Owen, Naomi C. Green, Daniel M. Espino, Duncan E. T. Shepherd. Visualization: Diana C. de Oliveira. Writing -original draft: Diana C. de Oliveira. Writing -review & editing: Diana C. de Oliveira, David G. Owen, Shuang Qian, Naomi C. Green, Daniel M. Espino, Duncan E. T. Shepherd. [/fig]
[table] Table 1: Catheter dimensions. Catheter tip designs A, B, C and D, with arterial and venous lumens indicated. Catheters A and B were set in reverse mode (for C and D designs, forward and reverse mode lead to the same model configuration). was evaluated for each boundary of the RA model (SVC, IVC and TV). The Reynolds number is defined as [/table]
[table] Table 2: Mesh settings and quality assessment. [/table]
[table] Table 3: Reynolds number study performed for the RA model. [/table]
[table] Table 4: Gauge pressure values applied at catheter outlets and respective flow rates achieved[5]. [/table]
[table] Table 5: CFD set-up. [/table]
[table] Table 6: Dimensions for the creation of tip volumes for all catheters.https://doi.org/10.1371/journal.pone.0247438.t006 [/table]
[table] Table 7: Haemodynamic predictions for all catheter models. [/table]
|
D-RNA Molecules Associated with Subisolates of the VT Strain of Citrus Tristeza Virus which Induce Different Seedling-Yellows Reactions
Citrus tristeza virus (CTV) strains were previously catalogued as seedling-yellows (SY) and non-SY (nSY) types, according to their yellowing and stunting effects on indicator seedlings. Among subisolates of the VT strain, which were selected from chronically infected Alemow plants, there was a correlation between the presence of 2.4-, 2.7-and 4.5-kb D-RNAs, and SY and nSY reactions, respectively. Similarly, plants infected with Mor-T subisolates, which cause SY, contained D-RNAs of 2.6 to 2.8 kb, while nSY subisolates from recovered sour orange tissue contained a major D-RNA of 5.1 kb. Plants harboring the 2.7-kb D-RNA were protected against challenge inoculation with a subisolate harboring the 4.5-kb D-RNA. This study suggests that the nSY reaction results either from the absence of SY gene(s) in the genomes of certain CTV strains or through the suppression of the effects of SY gene(s) by D-RNAs with 5 H parts larger than 4000 nt.
# Introduction
Citrus tristeza virus (CTV), a member of the closterovirus group and the Closteroviridae family (3±7) is an important pathogen, causing considerable economic losses to citrus industries worldwide. Citrus trees infected with CTV display two main types of disease: (i) quick decline of sweet oranges (SwO) (Citrus sinensis L.) and of some other species grafted on the sour orange (C. aurantium) rootstock; and (ii) stem pitting of grapefruit (C. paradisi) and pummelo (C. grandis). Other manifestations of infection with CTV include the seedling-yellows (SY) reaction (9±12) which is primarily a disease of experimentally inoculated plants but which might also be encountered in the ®eld in top-grafted plants. Seedlings of sour orange, lemon (C. limon) and grapefruit become chlorotic and stunted when inoculated with CTV-SY isolates, but no symptoms are elicited when SwO or mandarin (C. reticulata) is inoculated. The CTV-SY phenomenon is one of the long-standing enigmas in citrus virology. The early studies of McClean & van der Planck (9), Fraserand all suggested a complex aetiology of the CTV-SY disease. There have been reports of spontaneous recovery from SY infection by sour orange plants which initially showed SY symptoms, and of the elimination of the SY causal agent by the passage of SY-inducing CTV subisolates through SY-sensitive citrus hosts such as grapefruit and sour orange, which has led to the emergence of non-SY (nSY) isolates. These phenomena have given rise to the hypothesis that the CTV-SY reaction is caused by two separate components: the CTV agent, capable of autonomous replication and responsible for the quick decline and the lime reaction; and a second component, responsible for the SY reaction and able to replicate only in plants harboring the CTV component.
The CTV particles contain a single-component positive-stranded genomic RNA of 19296 nt for the Florida isolate, T36and of 19226 nt for the VT strain from Israel. The genomes of these CTV strains showed considerable sequence deviation within the 5 H half, but were found to have similar organization and to encompass 12 ORFs which potentially code for at least 17 protein products. In addition to the large replicative form (RF) RNA molecule, the infected plants contain a nested set of at least nine smaller species of 3 H -co-terminal single-and double-stranded subgenomic RNAs (sgRNAs). These sgRNAs correspond to the 3 H -terminal ORFs. Cloning of the VT strain of CTV revealed the presence of several defective (D) RNAs of various sizes, composed of the 5 H and 3 H termini of the genomic RNA with extensive internal deletions, along with the full-length virus. The sizes of the termini varied among species, with minimal lengths of 442 nt and 858 nt from the 3 H and the 5 H termini, respectively, resulting in different sizes of D-RNAs with different junction sites.
Inoculation of VT on the sour orange indicator resulted in SY symptoms. Later infections of sour orange seedlings by grafting with CTV-VT infected Alemow budwood resulted in inconsistent SY reactions; and not all plants showed the SY symptoms. Recently, we selected subisolates of two CTV strains, VT and Mor-T, which differed in their SY reactions on sour orange seedlings. The present paper reports the association of D-RNAs with 5 H termini larger then 4000 nt, with VT and Mor-T subisolates which do not elicit the SY reaction. D-RNAs may be involved in the long-standing enigma of the complex etiology of the SY-CTV reaction.
# Materials and methods
## Virus sources and propagation
The VT strain was originally isolated in 1970 from a SwO cv. Valencia tree grafted on sour orange. The tree showed advanced quick-decline symptoms. Inoculation of sour orange plants with the VT inoculum maintained in sour lime caused typical SY symptoms. Later passages of the VT strain from sour lime and Alemow plants to sour orange often resulted in inconsistent SY reactions: not all sour orange seedlings showed the SY symptoms, even when inoculum from a single Alemow plant was used to infect groups of plants from a single seed source (Bar-Joseph, unpublished). Subisolates of CTV-VT [fig_ref] Table 2: summarizes the differing SY reactions of sour orange seedlings graft-inoculated with eleven... [/fig_ref] were randomly selected in 1994 from chronically infected Alemow plants which had been graft inoculated several years earlier (1988 to 1994) with different passages of this strain. The VT subisolates were maintained in a propagation glasshouse with temperatures ranging between 15 and 35 C. The SY reaction was assayed by grafting chip buds from infected Alemow stems onto sour orange seedlings grown in a temperature-controlled glasshouse facility with incandescent illumination to complete 20 h of light, and two temperature regimes (TR) of 26/18 C or 29/21 C for the normal and the semi-warm TR, respectively. In both TRs the high and low temperatures were maintained for 8 and 12 h, respectively, and the adjustment from the high daytime to the low night time level and vice versa took 2 h. The SY reactions were recorded 8 and 16 weeks after inoculation, for the normal and semiwarm TR, respectively. The Mor-T isolate originated from a declining Minneola tangelo tree. The virus was propagated in Alemow and was used to inoculate a group of sour orange seedlings, some of which were inarched with the CTV-tolerant rootstock Go-Tou. Sour orange twigs and leaves showing SY and SY recovery, respectively, were used to infect sour orange and Alemow seedlings.
## Northern blotting
Double-stranded (ds) RNAs were isolated from 5±7 g of Alemow or sour orange tissues, according to Dodds and Bar-Joseph. The RNAs were separated by electrophoresis in formamide-formaldehyde denaturating, 1.1% agarose gels, prepared in MOPS buffer, transferred to Hybond N membranes. The hybridization probes consisted of a 611-bp and a 762-bp cDNA fragment from the 3 H and 5 H ends of CTV-VT genome, respectively. The DNA probes were either non-radioactively labeled using the Gene Images Random Prime Labeling Module Kit from Amersham or radioactively labeled with 32 P according to . RNA probes labeled with 32 32P-UTP were synthesized, with the Riboprobe System-T7 kit (Promega) according to the manufacturer's instructions, from cDNA fragments of 611 bp and 762 bp of the CTV-VT 3 H and 5 H ends, respectively, cloned in pGEM (Promega).
## Antibodies to ctv, elisa and western blotting
Antibodies for ELISA capture were prepared in sheep primed with recombinant CTV coat protein (rCTV-CP) antigen and boosted with a partially puri®ed CTV preparation. The second antibodies were obtained from egg yolks of chickens immunized with rCTV-CP. The ELISA procedure for CTV viral antigen quanti®cation in different tissues, which were soaked overnight in the antibody-coated ELISA wells, was according to Bar-Joseph et al..
## Rt-pcr ampli®cation restriction and sequencing analyses of cdna fragments from the vt5 and vt12 subisolates
The cDNAs were prepared from dsRNA templates of VT5 and VT12, with primers P1 and P2 for the ®rststrand synthesis, and primers P3±P4 and P5±P6 for nested and direct PCR ampli®cation [fig_ref] Table 1: Primers used for preparing cDNA fragments from two CTV-VT subisolates [/fig_ref]. The cDNA fragments were separated by electrophoresis on 1% agarose gel. The bands were excised from the gel and tested with the restriction enzymes, Sac I and Nsi I (Promega). For sequence analysis we used primers P7 and P9; P10 and P11; P10 and P8 to obtain three cDNA fragments located at ORF1 (1300±2486), ORFs 9 10(17260±17857) and ORFs 9 10 11(17260±18397), respectively. The cDNA fragments were cloned into the pUC 57/T (Fermentas) and sequenced from both sides by using Sequenase Version 2 from USB. Sequences of at least 150 bases were read from the 5 H and 3 H termini of each of the cDNA fragments. The dsRNAs from Alemow plants infected with two Mor-T subisolates, desig-nated #a and #b for SY-recovered and SY-reacting plants, respectively, were poly-A tailed and used for ®rst-strand cDNA synthesis with primer dT14V [fig_ref] Table 1: Primers used for preparing cDNA fragments from two CTV-VT subisolates [/fig_ref] and for second-strand synthesis with primers P9 and P8, for nested PCR ampli®cation of the viral 3 H and with primers P12 and AD for the viral 5 H . The cDNA fragments were separated by electrophoresis on 1% agarose gel, cloned into pUC 57/T (Fermentas). Sequencing from both sides of the 3 H fragments, was performed by using Sequenase Version 2 from USB and the 5 H sequence was determined with the aid of an automatic sequencing machine.
## Interference experiments between two vt subisolates harboring different d-rnas
Two groups of 9 month old Alemow seedlings were graft inoculated at heights of 25±30 and 30 cm, with two chip buds from Alemow plants infected with VT5 or VT12, respectively. Two weeks post-infection (wpi), the plants were pruned and allowed to develop two side branches. Tests for the presence of the speci®c D-RNAs were conducted after 10 wpi. The plants where challenged, 20 wpi by top grafting with stems infected with the reciprocal subisolates. Two lateral buds were allowed to sprout from each of the protected plants and leaf and stem bark tissue were tested for the presence of D-RNAs by Northern blotting.
[formula] P1 CGGTTGGCAGCAGAAGAC À 906±917 P2 GATGGACCTATGTTGGCCCCCCATAG À 19203±19227 P3 CAAATTCACCCGTACCCTCCGGAAATC 8±34 P4 AGCGAAGGATATCATCCA À 692±709 P5 TGGCGCATATGTTAATGC 18611±18628 P6 ATGGACCATTGTTGGCCCC À 19206±19227 P7 GAGGACGCCGACGTGTC À 2470±2486 P8 CTTCAGTGCTAGCTGTGTTG À 18377±18397 P9 GCTACGTTCGTCACGTATAC 1301±1321 P10 ATGCATATGAGCATTCGACGTCT 17260±17276 P11 GTCCGACTTCATAGAGTGTAC À 17837±17857 P12 GATGGGCACCGGAATGGC À 738±755 dT14V ATGACCAATCAGATGGCAC(T) 14 V* AD ATGACCAATCAGATGGCAC *V represents either A, C, or G (31). [/formula]
## D-rna molecules associated with ctv-vt subisolates
# Results
Biological Characterization of VT and Mor-T Subisolates
## Variation in d-rna composition of plants infected with vt and mor-t subisolates
Hybridization with an approximately 0.7-kb cDNA probe or riboprobe from the 5 H end of the VT genome with dsRNA extracts from Alemow plants, revealed the presence of the large RF and the low-molecularweight tristeza 5 H -corresponding RNA molecules (LMT)and D-RNAs. VT-subisolates 6±8 and 13, and 1, 5, 9, 10, with apparently similar SY reactions, showed the presence of two types of D-RNAs, of 2.4 kb and 2.7 kb, respectively. The three nSY subisolates showed the presence of a 4.5-kb D-RNA [fig_ref] Figure 1: Northern blot hybridization of dsRNA extracts from [/fig_ref] and [fig_ref] Table 2: summarizes the differing SY reactions of sour orange seedlings graft-inoculated with eleven... [/fig_ref]. The hybridization patterns of dsRNAs extracted from sour orange seedlings infected with VT subisolates VT12 (nSY) and VT5 (SY) are shown in [fig_ref] Figure 2: Northern blot analyses of interference between two CTV-VT subisolates containing different D-RNAs [/fig_ref]. Only weak or no hybridization signals of genomic and/or defective RNA could be located in bark and leaves from the sour orange plant which showed severe SY compared with those from the nSY plant. Hybridization of dsRNAs from Alemow plants inoculated with Mor-T subisolates #a1 (nSY) and #b1 (SY), showed the presence of major large (ca. 5.1 kb) and small (ca. 2.6 kb) D-RNAs respectively [fig_ref] Figure 1: Northern blot hybridization of dsRNA extracts from [/fig_ref]. One of the SY Mor-T subisolates #c1 showed only weak bands of D-RNA molecules compared with the nSY subisolate #e1, which showed the major D-RNA of ca. 5.1 kb [fig_ref] Figure 1: Northern blot hybridization of dsRNA extracts from [/fig_ref]. Sequence analyses revealed that SY subisolate #b1 contained two D-RNAs of 2634 and 2815 nt with junctions of their 5 H termini located at positions 1772 and 1521, whereas the nSY subisolate #a1, contained a major D-RNA of 5125 nt, with the junction of the 5 H terminus located at position 4376 [fig_ref] Figure 3: A [/fig_ref].
## Rt-pcr and sequencing analyses of vt subisolates
The hybridization with the VT 5 H probe with different VT and Mor-T subisolates suggested a close relationship between their genomic RNAs. In order to examine the genomic composition of the VT5 (SY) and the VT12 (nSY) subisolates, we compared the sequences of termini of their genomes by means of nested RT-PCR and sequencing analyses. Primers P1 and P2 were used for ®rst-strand cDNA synthesis and primers P3, P4, P5 and P6 [fig_ref] Table 1: Primers used for preparing cDNA fragments from two CTV-VT subisolates [/fig_ref] ampli®cation. The resulting cDNA fragments for both subisolates gave the expected lengths for the 5 H (8-709) and 3 H (18611±19227) ends of their genome. Restriction analysis of these products with SacI and NsiI gave restriction fragments of identical size (not shown). Sequence analyses of internal regions, at least 150 nt in length, of three cDNA fragments positioned at different regions of the VT genome ( positions 1300±2486, 17260±17857 and 17260±18397) did not reveal any sequence deviation between the products obtained from the dsRNAs of the VT5 (SY) and the VT12 (nSY) subisolates (not shown).
## For pcr
## Interference between small and large d-rnas
The possibility of interference between two VT subisolates, VT5 and VT12, harboring the 2.7-and the 4.5-kb D-RNAs, respectively, was tested in Alemow plants. The dsRNAs from plants which had ®rst received a protective inoculation with either the VT5 or the VT12 subisolate and were later challenged by top grafting with the reciprocal subisolate, were hybridized with the 5 H -speci®c probe. At 18 weeks post challenge inoculation (wpci), the basal parts of each combination had predominantly the D-RNAs of the protective isolate (not shown). Later tests at 41 wpci showed only the 2.7-kb D-RNA in the basal parts of plants protected with VT5 [fig_ref] Figure 2: Northern blot analyses of interference between two CTV-VT subisolates containing different D-RNAs [/fig_ref]. Plants protected with VT12 showed the presence of either a conspicuous or a weak band of the challenging 2.7-kb D-RNA in addition to the 4.5kb D-RNA [fig_ref] Figure 2: Northern blot analyses of interference between two CTV-VT subisolates containing different D-RNAs [/fig_ref].
Sour orange seedlings infected with Alemow tissues from the interference experiments, which harbored both the 2.7-and the 4. stronger ELISA titers and higher dsRNAs concentrations [fig_ref] Figure 2: Northern blot analyses of interference between two CTV-VT subisolates containing different D-RNAs [/fig_ref].
# Discussion
## Selection and characterization of vt and recovered mor-t subisolates
Biological and molecular characterization of 11 VT subisolates, which were randomly selected from chronically infected Alemow plants, revealed the presence of eight SY and three nSY subisolates. The VT subisolates caused similar symptoms and comparable ELISA reactions in Alemow plants (not shown). The virus titers were considerably higher in sour orange plants infected with nSY than in those infected with SY subisolates. These differences were consistent among plants which were maintained under different TRs [fig_ref] Table 2: summarizes the differing SY reactions of sour orange seedlings graft-inoculated with eleven... [/fig_ref]. Low virus titers or the absence of virus (indicated by negative reactions on indicator plants) in sour orange leaves and roots showing severe SY symptoms, suggest the possibility that the SY isolates emit a long-distance signal for a hypersensitive reaction. A similar situation has been previously observed in mature trees infected with CTV-Mor-T, where the collapse of the sweet orange/ sour orange combination often preceded the spread and redistribution of the virus towards the upper parts of the infected trees.
The profound differences among the sour orange reactions to the various VT-subisolates were associated with the presence of different major D-RNAs. The nSY subisolates, 3, 4 and 12, showed the presence of a major band of 4.5-kb D-RNA, whereas the eight SY subisolates, 1, 5±10 and 13, showed the presence of two smaller D-RNAs of 2.4 and 2.7 kb, with no apparent difference in the intensity of the SY reaction to subisolates which contained either of the smaller D-RNAs. Infection of sour orange with tissues from Alemow plants concomitantly infected with mixtures of VT5 and VT12 resulted in reactions ranging from SY to nSY, with virus titers depending on the relative concentrations of the 2.7-and 4.5-D-RNAs in the inoculum source.
Previously, we showed variations in the presence of the 2.4-, 2.7-and 4.5-kb D-RNAs in Alemow plants infected with budwood from a single VT-infected source plant. Differences in D-RNA populations might have accounted for the previously noticed inconsistencies in the SY reaction of sour orange plants infected with VT strain (Bar-Joseph, unpublished). The selection of VT subisolates which show a more consistent SY reaction was correlated with the presence of a major type of D-RNA [fig_ref] Table 2: summarizes the differing SY reactions of sour orange seedlings graft-inoculated with eleven... [/fig_ref]. One probable reason for obtaining apparently stable subisolates was their selection from chronically infected plants ( 4 2±3 years after inoculation) at a time when a single type of D-RNA had become dominant. [fig_ref] Figure 3: A [/fig_ref]. A 16nt sequence, 5 H -GAAAACTAATTTATCA, with no homology to other regions of the CTV genome was found at the junction site [fig_ref] Figure 3: A [/fig_ref]. A different short sequence, probably of host origin had previously been observed at the junction site of the 2.4-kb D-RNA.
## Interference between vt subisolates harboring different d-rnas
The CTV-SY phenomenon is one of the longstanding enigmas in citrus virology. The ®nding that both the CTV and the CTV-SY diseases could be transferred by mechanical inoculation of preparations of CTV particles (26,27) raised the question (28) of the dual-component theory of the causal agent of the CTV-SY disease. Dodds et al.noted an association between two dsRNAs of about 0.8 and 2.7 kb and SwO trees infected with SY subisolates. Molecular characterization associated the 0.8-kbp dsRNA with the replicative subgenomic RNA coding for ORF11 (940 nt)and hybridization with a 3 H -speci®c probe did not reveal quantitative differences in the amounts of the 0.8kbp dsRNAs from SY and nSY plants (not shown). Moreover, low-molecular-weight D-RNAs of 2.4 kb were located in Alemow infected with nSY isolates Mik-T and Ach-T (32) (not shown).
CTV isolates were previously classi®ed by a variety of criteria into subisolates which differed in host reactions, vector transmissibility and dsRNAs patterns (29,33±37). The variability among subisolates was considered as an indication of the high frequency of mixed CTV infections. D-RNAs were previously implicated in the variability between the dsRNA patterns of parental isolates and their subisolatesand the present ®ndings indicate a correlation between certain D-RNAs and host reactions, and support a working hypothesis that the nSY reaction results either from the absence of SY gene(s) or through the suppression of their effects by D-RNAs with 5 H parts larger than 4000 nt.
The genomic and D-RNA fragments of the two differentially reacting VT subisolates were found to show a complete sequence identity. Nevertheless, the possibility that a minimal sequence deviation between other parts of their genomes is involved in these biological differences cannot at the present be completely ruled out. Moreover, the question of the mechanism that causes SY symptoms in sour orange tissues, which contain only low concentrations of viruses or D-RNA remains to be answered. D-RNAs have been isolated from a broad spectrum of animal viruses and, more recently, also from a large number of plant viruses (for recent reviews, see. Different D-RNAs have previously been reported to have different effects on disease expression: while D-RNAs of tombusviruses had attenuating effects on infection, the D-RNAs associated with the turnip crinkle virus tended to increase the severity of symptomsand the D-RNAs associated with broad bean mottle virus had no effect on some host plants but intensi®ed the severity of symptoms in others. The correlation between the SY reactions of sour orange seedlings and the genomic composition of the D-RNAs in the Alemow inoculum, support the notion that the host type is a major determinant of the biological effects of D-RNAs.
[fig] Figure 1: Northern blot hybridization of dsRNA extracts from: A. Alemow plants infected with different subisolates of the VT strain of CTV, using a non-radioactive cDNA probe from the distal 5 H end of the CTV genome (Lanes 1 to 9 represent subisolates 12, 5, 9, 10, 6, 4, 3, 1 and 13, respectively. B. Sour orange plants infected with VT subisolates 12 and 5 (lanes 1 and 2, respectively) and with inoculum from an Alemow plant which was included in the interference experiments, showed a dominant 4.5-kb and a minor 2.7-kb D-RNAs (lane 3). C. Alemow plants inoculated with Mor-T subisolates #a1 (nSY) and #b1 (SY) (lanes 1 and 2, respectively) and with subisolates #e1 (nSY) and #c1 (SY) (D. Lanes 1 and 2, respectively). [/fig]
[fig] Figure 2: Northern blot analyses of interference between two CTV-VT subisolates containing different D-RNAs. Lanes 1 and 2 show hybridization with dsRNA extracts from Alemow plants infected with VT subisolates 12 and 5. Lanes 3, 4 and 6 represent plants protected with VT5 and challenge inoculated with VT12, at 41 weeks post challenge inoculation. Lanes 5 and 7 represent plants protected with VT12 and challenged with VT5. Note the absence of the D-RNA harbored by the challenging subisolate in plants protected with the smaller (2.7 kb) D-RNA (lanes 3, 4 and 6) and the considerably larger amounts of the challenging (2.7 kb) D-RNA in lane 7 than in lane 5. [/fig]
[fig] Figure 3: A. Diagram of the genomic organization of CTV. B. The structure of the three CTV-Mor-T-D-RNAs compared with the CTV genomic RNA. Note the presence of a 16-nt sequence apparently of non-viral origin at the junction site of the 5.1-kb D-RNA. [/fig]
[table] Table 1: Primers used for preparing cDNA fragments from two CTV-VT subisolates [/table]
[table] Table 2: summarizes the differing SY reactions of sour orange seedlings graft-inoculated with eleven CTV-VT subisolates. Plants infected with VT subisolates 1, 5±10 and 13 showed typical SY symptoms, including short internodes and thorns and small leaves (not shown). Plants infected with VT subisolates 3, 4 and 12 developed similarly to non-infected plants or, occasionally, showed a few smaller leaves and were de®ned as nSY. The root systems of the SY-reacting sour orange plants were especially damaged and reduced in weight, compared with the control noninfected or the nSY plants (not shown). Among the plants with SY symptoms which were maintained under the high TRs, a previously unrecorded symptom, midday wilting, was noticed 5±6 weeks after infection (not shown).The virus titers estimated by ELISA, of sour orange plants maintained under two TRs after infection with various VT subisolates are presented in. The virus titers under both TRs were signi®cantly higher for the nSY than for the SY subisolates (). The virus titers of the apical parts() and of the root and rootlets (not shown) of plants infected with VT subisolates 1, 7, 8 and 10, which showed wilting and SY reactions under the higher TR, were either similar to or only slightly above those of the healthy controls. Moreover, grafting of apical sour orange leaves with severe SY symptoms on Alemow plants resultedin 4/18 of virusfree plants. Contrary to these profound differences among the sour orange plants, the ELISA titers of Alemow plants infected with VT5 or VT12 were similar (not shown). Infection of the Mor-T#o isolate on sour orange plants caused a severe SY reaction in all (n 6) infected plants. These symptoms persisted for at least 18 mo, except for the plants which were inarched with the Go Tou rootstock. In the inarched plants, apparently recovered twigs with larger leaves were observed after 1±2 years. ELISA assays showed higher titers for the SY-recovered (SYr) parts compared with the SY leaves (not shown). The sour orange plants n 4 infected with stem pieces from the SY part showed typical SY symptoms. The sour orange plants infected with SYr leaves showed different symptoms, with 2, 2 and 1 plants showing nSY, SY and intermediate reactions respectively. Virus titers estimated by ELISA, in sour orange petioles were 4 1.0 or 5 0.2 OD/405 nm for plants with Mor-T nSY and SY subisolates, respectively. [/table]
|
Pilot phase of an internet-based RCT of HIVST targeting MSM and transgender people in England and Wales: advertising strategies and acceptability of the intervention
Background: The SELPHI study (An HIV Self-Testing Public Health Intervention) is an online randomised controlled trial (RCT) of HIV self-testing (HIVST). The aim of this study was to assess the feasibility of recruiting UK men who have sex with men (cis and trans) and trans women who have sex with men to the SELPHI pilot, and the acceptability of the HIVST intervention used among those randomised to receive a kit.Methods: A mixed-methods approach to assessing trial feasibility and intervention acceptability was taken, using quantitative data from advertising sources and RCT surveys alongside qualitative data from a nested sub-study. Results: Online recruitment and intervention delivery was feasible. The recruitment strategy led to the registration of 1370 participants of whom 76% (1035) successfully enrolled and were randomised 60/40 to baseline testing vs no baseline testing. Advertising platforms performed variably. Reported HIVST kit use increased from 83% at two weeks to 96% at three months. Acceptability was very high across all quantitative measures. Participants described the instructions as easy to use, and the testing process as simple. The support structures in SELPHI were felt to be adequate. Described emotional responses to HIVST varied.Conclusions: Recruiting to a modest sized HIVST pilot RCT is feasible, and the recruitment, intervention and HIVST kit were acceptable. Research on support needs of individuals with reactive results is warranted.
reducing stigma and privacy concerns as well as increasing convenience and ameliorating geographical barriers in areas underserved by other HIV testing opportunities [bib_ref] Attitudes and acceptability on HIV self-testing among key populations: a literature review, Figueroa [/bib_ref] [bib_ref] Supervised and unsupervised self-testing for HIV in high-and low-risk populations: a systematic..., Pai [/bib_ref]. It also provides flexibility in intervention design: components and delivery mechanisms can be adapted depending on the target population [bib_ref] Attitudes and acceptability on HIV self-testing among key populations: a literature review, Figueroa [/bib_ref]. In 2016 the World Health Organization incorporated HIVST into its Consolidated Guidelines for HIV Testing Services with the recommendation that HIVST be provided as a supplementary option alongside existing services.
Evidence from UK studies conducted shortly after HIVST became commercially available in 2015 suggests that HIVST is acceptable, and that MSM have preferences for blood-based tests (due to perceptions of greater accuracy) and easy access to confirmatory testing [bib_ref] HIV testing history and preferences or future tests among gay men, bisexual..., Witzel [/bib_ref] [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref] [bib_ref] Preparedness for use of the rapid result HIV self-test by gay men..., Flowers [/bib_ref]. However, a minority of MSM with aversion to blood reported being unwilling to use a blood-based kit [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref]. Home delivery of an HIVST kit is a barrier for some with concerns around domestic privacy [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref]. HIVST may be more appealing to groups who do not test in line with current guidelines, which recommend annual testing for all MSM, or more frequent testing if at increased risk [bib_ref] HIV testing history and preferences or future tests among gay men, bisexual..., Witzel [/bib_ref]. Evidence also suggests that challenges relating to instructions and lack of familiarity with testing procedures may present barriers to use, particularly initially [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref]. Little data exists on trans populations, although a study in San Francisco found HIVST was acceptable and feasible for trans women [bib_ref] Acceptability and feasibility of HIV self-testing among transgender women in San Francisco:..., Lippman [/bib_ref]. In addition, a small number of trans women have accessed England's national HIV self-sampling (HIVSS) service indicating that testing outside clinics may be preferable to some.
HIVSS, whereby a person takes their own sample and returns it to a lab that then processes it and provides a result, is the technology perhaps most analogous to HIVST. HIVSS has suffered from sub-optimal sample returns, with testing completion rates around 55% in service evaluations in the UK [bib_ref] Identifying undiagnosed HIV in men who have sex with men (MSM) by..., Elliot [/bib_ref]. Evidence suggests this relates to complicated sampling procedures which are not always feasible or acceptable to the target populations, including taking a sufficiently large blood sample to facilitate testing [bib_ref] Feasibility and acceptability of self sampling kits to increase the uptake of..., Seguin [/bib_ref] [bib_ref] Acceptability of HIV self-sampling kits (TINY vial) among people of black African..., Dodds [/bib_ref]. If those accessing HIVST face similar barriers in performing tests this could threaten the aspiration of increasing testing uptake and frequency through the provision of this novel technology. Further, although the expansion of commercially available HIVST has seen moderate levels of uptake in the USA [bib_ref] HIV self-testing: a review of current implementation and Fidelity, Estem [/bib_ref] , HIVST may fulfil different roles for populations such as in the UK where HIV testing services are very well developed. This is especially true should HIVST also be provided at no cost, as with the vast majority of existing HIV testing models in the UK.
The SELPHI study (An HIV Self-Testing Public Health Intervention) is an online randomised controlled trial (RCT) being conducted between 2017 and 2020 which aims to assess whether HIVST can; (i) increase rates of diagnosis in those with prevalent HIV infection and ii) reduce the time between infection and diagnosis for those at risk of incident infection. The primary outcomes are ascertained through linkage to the national HV surveillance systems indicating confirmatory testing and linkage to care. SELPHI aimed to recruit 10,000 MSM (cis and trans) and trans women. Participants were recruited through advertising on geo-location social-sexual networking applications and Facebook. Initial baseline randomisation was to an offer of postal delivery of an HIVST kit accompanied with a follow-up survey or to no HIVST.
Intervention conceptualisation was underpinned by the COM-B model of behaviour change [bib_ref] Validation of the theoretical domains framework for use in behaviour change and..., Cane [/bib_ref] [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref]. COM-B is a systematically developed model which consolidates 19 pre-existing frameworks, positing that alterations in capability, opportunity and motivation are key to successful behaviour change interventions [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref]. This model was chosen because of its simplicity and flexibility, and because of its use in HIV prevention interventions as well as interventions which include the provision of technologically assisted behaviour change [bib_ref] Applying the COM-B behaviour model and behaviour change wheel to develop an..., Barker [/bib_ref] [bib_ref] Interactive two-way mHealth interventions for improving medication adherence: an evaluation using the..., Chiang [/bib_ref] [bib_ref] Development of a theory-based interactive digital intervention to improve condom use in..., Webster [/bib_ref] [bib_ref] The use of smartphone health apps and other mobile h ealth (mHealth)..., Chen [/bib_ref] [bib_ref] Feasibility and acceptability of home sampling kits to increase the uptake of..., Seguin [/bib_ref]. The pilot phase of SELPHI ran from February to May 2017 and aimed for 1,000 recruits, from the overall target of 10,000.
Evidence on HIVST intervention implementation feasibility and acceptability in high income settings to date has focused on small scale demonstration studies distributing small numbers of kits with limited follow-up [bib_ref] Access to HIV self-testing doubles the frequency of HIV testing among gay..., Jamil [/bib_ref] [bib_ref] New initiatives to develop self-testing for HIV, Witzel [/bib_ref]. The SELPHI pilot provides an opportunity to generate evidence about whether large-scale implementation of an online HIVST RCT is feasible in high-income settings. Usability of HIVST, defined as "the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use", can also be assessed. It is also vital to understand intervention acceptability among those who receive the intervention to inform practitioners, policy makers and commissioners. This study, which is grounded in implementation science, will be useful in a range of contexts with similar health system features and HIV epidemics.
The aim of this study is therefore to assess the feasibility of recruiting to an online HIVST RCT in which participants are randomised to receive a free kit or not, and the acceptability of the HIVST intervention used among those randomised to receive it. We consider key questions related to advertising performance, reach, uptake, kit usability and end user reception. We use a mixed methods approach examining the feasibility of recruitment, the motivations of SELPHI participants, and the usability and acceptability of the kit itself. Theoretically, this work is informed by COM-B, a behaviour change model which is often used to explore acceptability and to conceptualise intervention components and how they may work together to produce behaviour change [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref] [bib_ref] Applying the COM-B behaviour model and behaviour change wheel to develop an..., Barker [/bib_ref] [bib_ref] Efficacy of behavioural interventions for transport behaviour change: systematic review, meta-analysis and..., Arnott [/bib_ref].
# Methods
This mixed-methods study follows a data integration approach termed by Moran-Ellis et al. as following a thread [bib_ref] Triangulation and integration: processes, claims and implications, Moran-Ellis [/bib_ref]. As such key areas of inquiry were identified from quantitative RCT data; these were then used to guide the focus of the analysis of the in-depth interviews. This has allowed us to generate additional nuance in responding to questions about feasibility and acceptability.
## Rct study procedures
The pilot was designed to test the RCT recruitment strategy and the procedures in place for the full online trial. The pilot also tested the likely acceptability of the intervention, especially the usability of the chosen HIVST kit, its delivery mechanisms and the support offered for its use. Full details of the RCT methods can be found in the published protocol [bib_ref] Protocol, rationale and design of SELPHI: a randomised controlled trial assessing whether..., Gabriel [/bib_ref].
Eligible participants were men (cis and trans) and trans women; reporting lifetime anal sex with a man; not known to be HIV positive; aged 16 years and older; resident in England or Wales; willing to provide name, date of birth, postal and email address; consent to linkage with surveillance and clinic databases and not previously enrolled to the study.
The recruitment strategy utilised adverts placed in geo-location social-sexual networking applications (apps) (Grindr, Growlr, Scruff & Hornet) as well as targeted Facebook advertising. Free advertisements were placed on the Facebook page of a transgender focused clinical service. Recruitment sources were chosen based on previous experience, and through consultation with voluntary sector organisations. Grindr was chosen as it has the largest market share in the UK, with Hornet targeting a similar group. Growlr caters to a largely older subgroup of MSM, while Scruff is ostensibly most used by hirsute MSM and their admirers. Some adverts targeted a national audience, while others took a city or regional approach. Messaging was devised drawing learning from earlier formative work [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref] [bib_ref] Risk, reassurance and routine: a qualitative study of narrative understandings of the..., Witzel [/bib_ref] , and with participant and public involvement (PPI) representatives. Key themes regarding barriers and facilitators to recruitment were identified, and two members of the study team met with PPI co-chairs to develop specific advertising messages. Adverts focused on all COM-B domains: capability was addressed through promoting ease of HIVST use; opportunity was addressed through highlighting the HIVST kits were available at no cost; and motivation was enhanced through highlighting privacy and appealing to altruism to take part in a study. Some messages specifically highlighted trans eligibility. Advertisements appeared as sponsored posts, as direct inbox messages, as pop-up messages and as banners.
Participants were directed to a registration survey requiring informed consent and confirming eligibility, and then to an enrolment survey via email. Ineligible participants (and those not randomised to HIVST) were offered additional information on HIV testing. The enrolment survey asked additional demographic and behavioural questions. Participants were randomised 60:40 to receive an HIVST kit (baseline test [BT]) or to no kit offer (no baseline test [nBT]). Kits were distributed by post, directly to the given address by the test manufacturer (BioSure™).
Two weeks after enrolment, participants randomised to receive the HIVST kit were emailed an online followup survey asking if the kit had been used (and if not, why not), what the test result was, and whether further care was accessed. Two reminders were sent.
Three months after randomisation a survey was emailed to all participants asking for information on testing and risk behaviour in the intervening period. Two reminders were sent. Participants randomised to BT were also asked questions about their experiences with HIVST. They ranked on a 5-point scale their agreement with statements related to acceptability and usability of the kit: 1) the instructions were easy to use; 2) performing the test was simple; and 3) my overall experience was good.
## Intervention development
The intervention being trialled was linear. The recruitment messages being tested were both part of the intervention and trial process, but in a scaled-up intervention delivering HIVST these would be adapted accordingly. A brief HIV risk assessment was conducted through behavioural questions in the enrolment survey. The kit and accompanying sleeve were then delivered and two weeks later a follow-up survey asked about kit use and the test result. Those who reported not receiving a kit had a new delivery arranged. These components (advertisement, risk assessment, kit and two-week follow-up) were defined as the intervention as all were theorised to increase engagement with HIV testing through COM-B channels.
Formative work was central to intervention development. Focus groups with MSM and key informant interviews identified specific barriers to uptake and use of the HIVST which the SELPHI intervention development was attentive to ameliorating, using COM-B. These efforts were also used in developing appropriate messaging for advertising. [fig_ref] Figure 1: Intervention diagram with COM-B categorisation and targeted domains [/fig_ref] provides a visualisation of intervention components with a description of the intervention functions and the COM-B domains they seek to affect.
Anticipated concerns regarding ease of use, coded in COM-B as capability (physical), were addressed in advertisements (see [fig_ref] Figure 2: Advertising samplesDemographic features of sample including by recruitment source [/fig_ref] for examples). This combined intervention approaches described as persuasion and education in COM-B [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref] , enhancing motivation by minimising concerns regarding ease of use and highlighting privacy and convenience. Issues concerning lack of knowledge in using HIVST were also identified.
The COM-B model codes this as psychological capability; the behaviour change wheel then suggests intervention functions such as education, training and enablement might be useful [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref]. This was alongside an observed preference for additional supportive information beyond what was provide in the original BioSure™ kit. This necessitated the development of a sleeve over the box holding the kit to provide support information (education), as well as behavioural support (enablement) which was highlighted in the two-week follow-up survey. The sleeve also provided signposting to a free telephone helpline and website for HIV and sexual health information (enablement).
In order to increase engagement with HIV testing generally, a risk assessment was included in the enrolment survey to provide a reflective experience examining personal risk. It was theorised that this approach (persuasion) can increase motivation (reflective and automatic) [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref].
Formative research also identified issues with the instructions and packaging of an earlier iteration of the kit, both of which reduced motivation to access HIVST and capability when doing so. The kit instructions were reformatted by the manufacturer before implementation began, effectively addressing this issue. This intervention component is theorised as training (the imparting of skills) in the COM-B system [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref].
The broader HIVST literature identifies support issues as a key concern in HIVST delivery, a concern also identified in our formative work [bib_ref] Attitudes and acceptability on HIV self-testing among key populations: a literature review, Figueroa [/bib_ref] [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref] [bib_ref] A review of the evidence of harm from self-tests, Brown [/bib_ref]. This informed the provision of enhanced support information via our kit sleeve produced in collaboration with our community advisory group co-chairs (see Additional file 1). The two-week follow-up survey was also designed to counter this concern. If a participant reported a positive result here, they were directed to a page providing information on how to find their local HIV clinic. This same page was linked to from the three-month survey. Additional information about receiving a positive result was provided on the SELPHI website.
## Data handling, generation & analysis
Data pertaining to advertising reach was recorded for all adverts, then pooled according to platform. The click conversion rate (proportion of those clicking on the advert who subsequently registered) was calculated. Eligible and ineligible registrations as well as the number of successful randomisations were tabulated. Registration conversion was calculated by deriving the proportion of eligible registrations who filled in the enrolment survey and were subsequently randomised.
Baseline demographic and behavioural profiles were tabulated overall and by recruitment source. Variables considered were age (both continuous and 10 year bands), gender, sexual orientation, ethnicity (recoded from standard UK ethnicity codes into white, Asian, black & other), highest educational qualification (low: GCSEs and below; medium: A-levels or equivalent, higher education below degree level; high: degree or higher), HIV testing history (tested in preceding 12 months; tested more than 12 months ago; never tested), and condomless anal intercourse (CAI) in the preceding 3 months. Participant demographic and behavioural characteristics were compared between recruitment sources using chi-squared tests or a Kruskal-Wallis test for age.
Responses to the 2-week survey were summarised by proportion who completed the survey, proportion who received the kit, and the proportion who subsequently used the kit.
Kit use was summarised again from the 3-month survey alongside acceptability variables pertaining to instructions, simplicity of test performance and overall experience.
## Qualitative data
A qualitative study was undertaken with 10 cis-gender MSM participants during the pilot in order to examine intervention acceptability in greater depth. Participants were sampled purposively from those randomised to receive an HIVST kit. Sampling aimed to be diverse with regard to testing history: whether an individual had tested in the 12 months before joining SELPHI; not tested in the preceding 12 months; or never previously tested for HIV. Efforts were made to ensure sample diversity with regard to demographic features, especially education.
A semi-structured interview topic guide was developed to explore questions from formative research [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref] [bib_ref] Risk, reassurance and routine: a qualitative study of narrative understandings of the..., Witzel [/bib_ref] , including issues related to capability, HIVST potential, anticipated responses and acceptability and mapped onto COM-B. The guide covered HIV testing history, motivations for joining and experiences of the SELPHI RCT, questions related to using HIVST and emotional responses.
Interviews were conducted over the phone or through Skype, and participants were electronically given a £30 incentive. All interviews were audio recorded and transcribed.
A thematic framework was developed for analysis, fusing the approaches described by Braun and Clarkeand Richie and Spencer [bib_ref] Qualitative data analysis for applied policy research, Ritchie [/bib_ref]. This inductive process involved familiarisation with the transcripts and drawing out emerging themes. These themes were arranged into groups, with higher-level themes emerging from subthemes, both organised hierarchically, and again mapped onto COM-B to better elucidate how acceptability of intervention components related to the behaviour change domains. The framework was piloted on two transcripts, refined, and applied to all remaining transcripts. We draw data from across this framework and report themes by COM-B domain for simplicity of interpretation.
# Results
The recruitment strategy led to the registration of 1370 eligible participants through 13 advertisements across 5 platforms, of whom 76% (1035) subsequently enrolled and reached baseline randomisation. In this pilot, 631 participants were randomised to receive an HIVST kit (BT), while 404 were randomised to not receive a kit (nBT). Of those randomised to BT, 66% (415) completed the two-week follow-up survey and 64% (405) completed the first three-month survey. Overall 78% (494/631) completed at least one of these two surveys (2-weeks or 3-months).
## Recruitment strategy performance
Click conversion was highly variable, from 8% in Grindr adverts to 20% in Facebook advertising. Registration conversion ranged from 71 to 80% (mean = 76%). Cost per randomised participant varied: Hornet was cheapest (£1.66) and Grindr most expensive (£7.16). Costs were stable through this phase, with no evidence in the pilot of diminishing returns. See [fig_ref] Table 1: Advertising source data [/fig_ref] for full details. [fig_ref] Table 2: Participant demographics by advert source and overall IQR interquartile range, CAI condomless... [/fig_ref] presents baseline demographics overall and by recruitment source. [fig_ref] Figure 3: Map [/fig_ref] presents the geographic distribution of randomised participants with each dot representing a group of multiple randomised participant from a source, coded by colour. The recruitment strategy engaged a range of MSM, but less so trans women. Median age was 32.1 years (IQR 25.9, 41.6). Cis-gender MSM comprised the majority (99%) of the sample, as did participants of white ethnicity (89%, n = 921), and MSM who identified as gay (89%, n = 757). Most (60%, n = 611) were highly educated and reported CAI within the preceding 3 months (70% n = 726). Sixty-four percent of participants (n = 652) had tested for HIV in the preceding year and 14% (n = 141) had never previously tested. Of never tested participants, 82 (58%) reported one or more CAI partners in preceding 3 months. [fig_ref] Table 2: Participant demographics by advert source and overall IQR interquartile range, CAI condomless... [/fig_ref] presents full details of baseline demographics.
There were significant differences in age (p < 0.001) and gender (p = 0.01) across recruitment sources, with all other variables being similar. As anticipated, Growlr recruited older participants whereas Facebook recruited younger ones. Free advertising targeted towards trans people was most effective for reaching trans participants, although numbers were small. See [fig_ref] Table 2: Participant demographics by advert source and overall IQR interquartile range, CAI condomless... [/fig_ref]. [fig_ref] Table 3: Qualitative sub-study sample [/fig_ref] presents demographic characteristics of the qualitative sub-study. All participants were cis-MSM. When discussing their motivations for joining SELPHI, participants described three predominant motivations: i) to access HIV testing; ii) desire to use a novel technology; and iii) altruism.
## Motivations of participants (qualitative sub-study)
## Accessing testing
HIVST reduced specific HIV testing barriers, thereby facilitating uptake. This was especially true for those who had never tested, and those who had not tested within the preceding twelve months. Opportunity barriers (e.g. convenience and ease of access) and motivational barriers (e.g. confidentiality and stigma) were ameliorated by HIVST.
Sometimes people ask you what you're coming in for, [ … ] if you say, 'oh I'm coming for an HIV [test]' they think you're gay, or they think you're disgusting, you don't use protection, or blah, blah, blah. But it's mainly about being labelled as something you're not. (23-year old bisexual man, never tested).
## Desire to use a novel technology
Just under half of those interviewed reported being motivated to join SELPHI out of a desire to experience a novel technology or because they felt SELPHI was a new kind of study. HIVST was understood to be an evolution of HIV testing methods which was appealing to some:
It's an interesting one because it's obviously very new. So you kind of think, well it's really great.
[ … ] You just think it's something that's interesting to try because it's new technology. year old gay man, tested in last 2 years).
## Altruism
Altruistic motivations were reported by just over half of participants with a range of testing histories. These were predominantly secondary motivations, helping support the decision to join a trial. Motivations were related to notions of good citizenship, desire to contribute to the gay community and to science more broadly.
I find it quite interesting actually that those kinds of services are targeted towards people through Facebook because you're kind of transpiring an audience of people who might benefit from that service. And I thought, Registration conversion: proportion of eligible registrations leading to a randomisation "Actually that's quite smart" because I'm in that audience. And so, I just thought, "Yeah, I will give it a try" year old gay man, tested in last 3 months).
## Kit use at two weeks and three months
Of 631 who were randomised to BT, 66% (415) completed a two-week follow-up survey. At this point, 95% (394) reported having received their kit and 83% (328) of those had used it themselves. Reasons for not using the kit were mainly that participants were planning to use it in the future (97% n = 64) or that participants had tested elsewhere instead (3%). At the three-month survey, completed by 64% of eligible participants, 97% (390/403) reported having received, of which 96% (375/390) had used the kit. This indicates that although a significant minority delayed kit use, most did use the test kit by three months.
When results from both surveys were pooled, providing data for 78% of participants, 97% (477/494) received the kit and 90% (445/494) had used it. Assuming that all those participants that did not complete either of these surveys received the kit (137/ 631), but none used it, then the lowest possible estimate of kit use was 71% (445/631).
## Hivst usability and acceptability
At three months, participants reported very high HIVST usability and acceptability. Of 375 who used the kit and completed the three-month survey, 98% (362/369) found [fig_ref] Figure 4: Intervention acceptability [/fig_ref]. All qualitative interview participants had used their kits to test themselves. Below we describe intervention acceptability as it relates to the main domains of COM-B: capability, opportunity and motivation.
## Capability (physical & psychological)
Capability was the most pronounced of the three COM-B domains in our acceptability analysis, especially around test kit usability. Themes around physical capability tended to concern the instructions and using the lancet to take a blood sample. The inclusion of the two-week follow-up processes was a valued intervention element addressing psychological capability.
The instructions were generally felt to be easy to understand and interpret, although one participant felt they did not cater to a sufficiently diverse range of skills. The testing process was described as simple and the result was easy to interpret: Very clear and it was quite obvious as well what goes where and how to do it. It was clear. The descriptions and the pictures were easy to follow (29 year-old man, undisclosed sexual orientation, tested in last 5 years).
Blood collection via the included lancet was a barrier for some. Those who had no previous experience of drawing blood with a lancet reported concerns about their capability to collect their own sample, although all felt with experience this would no longer be an issue.
I actually don't like getting my finger pricked [ … ] so I was most worried about the finger prick, [ … ] so for me that was the most difficult thing, and then I wasn't sure if I was getting enough blood [ … ] but once I pricked the finger and I collected the blood then it was pretty straightforward (31-year old gay man, tested last 4 weeks).
Capability (psychological) also emerged when discussing the support components of the intervention. Participants generally felt that the supporting information provided was adequate and did not diminish acceptability of the intervention.
What I think relieved me of most of the anxiety was actually the kit included a card saying, at the end, if you are diagnosed with HIV then not to worry, here's what you can do. A, B, C. And if you're not, great. A, B, C. And I think the steps on that card saying, if you are, step one, step two, step three, it helped to relieve some of the uncertainty of what might happen if the test came out positive. (16-year old gay man, not previously tested)
## Opportunity (physical & social)
Themes related to acceptability of the entire intervention package were primarily related to opportunity (physical and social). For participants located in areas underserved by HIV testing opportunities, the kit ameliorated geographic barriers. Individuals who faced psychosocial barriers to testing felt HIVST gave them increased privacy around testing, enhancing the acceptability of the intervention.
I'm quite a private person. I like to keep certain aspects of my life to myself and sometimes people might be bothering you to talk about things where you think, "Well, I'm not there yet." [ … ] Whereas I can let that sink in and think, "Right, okay, now I'm ready to go and do whatever I need to do or talk to whoever I need to talk to." (34 year-old gay man, tested more than 12 months ago).
## Motivation (reflective & automatic)
The dislocation of HIVST from care pathways affected acceptability through motivational channels. Despite high acceptability related to the follow-up provided, HIVST as a concept was perceived to be associated with increased anxiety relative to other testing opportunities. This was largely due to concerns about conducting a test alone, and the potential separation of initial "diagnosis" from established care pathways.
[ … ] I think slightly the kit at home [makes me more anxious]. It's almost because it's literally taken out of your hands when you go to an STI clinic. So you don't have to think about it as much. It's something done to you. year-old gay man, tested within last 5 years) One participant delayed using the kit due to anxiety and instead visited a GUM clinic, saving his HIVST for use at a later date.min interval between conducting the test and reading the result was described as an exceptionally anxious time, especially for those who had tested due to risk.
I was feeling nervous actually because I was thinking what about if it does come back positive. That was quite like a bit of a head scratcher, waiting for the 15 minutes, and then when 15 minutes were up and I looked at the result and I was like, oh, you know it was negative so I thought right, but then I thought well yeah, maybe I shouldn't have worried that much, but you can't help it (23-year old bisexual man, not previously tested).
All participants described significant relief when reading a negative (non-reactive) result. A minority, who had more experience of HIV testing, felt that HIVST was associated with less anxiety than testing methods relying on laboratory-run tests due to the relative immediacy of results.
# Discussion
Through this mixed methods study we assessed the feasibility of recruiting MSM and trans people to the pilot phase of the online SELPHI RCT and the acceptability of HIVST, focusing mainly on the acceptability of the intervention and the usability of the kit.
Advertising performance varied according to platform by click and registration conversions and, crucially, cost. The pilot sample was predominantly white, well-educated, gay identified cis-MSM who reported CAI in the 3months preceding and who had tested for HIV in the preceding 12-months. The pilot struggled to recruit significant numbers of trans people, particularly trans women. Our recruitment did however, reach a range of participants across demographic groups. Platforms recruited participants of a similar demographic and behavioural profile except when considering age and gender identity.
Sixty-six percent of participants completed the twoweek follow-up, and 64% the three-month survey. Overall 78% of participants randomised to receive HIVST completed at least one of the two. Kit use was high, increasing from 86% at two-weeks to 96% at three months. The lowest possible estimate of kit use was 71%, assuming that all those not completing the follow-up surveys did not use their kits, which is unlikely. The kit was considered usable and the intervention was acceptable across the three dimensions interrogated (ease of use of instructions, test simple to perform and overall experience). Qualitative data provides nuance, with some participants reporting difficulty using the lancet. The relationship between HIVST and anxiety was ambiguous; individuals thought it could increase or ameliorate anxiety depending on previous HIV testing experience. Our recruitment strategy was successful in reaching a group at risk of HIV who had not HIV tested previously, with 58% of never tested MSM reporting CAI in the 3months preceding their enrolment. This is a key group with clear HIV prevention needs who should be a primary target for new testing interventions. The sample recruited in the pilot was comparable to previous convenience samples of MSM, as well as the ethnic makeup of the UK [bib_ref] Men who have sex with men in Great Britain: comparing methods and..., Prah [/bib_ref]. This indicates that this type of recruitment strategy is capable of reaching a group broadly representative of UK MSM in terms of ethnicity. A group which is underrepresented when compared to national statistics is MSM of Asian ethnicity. This could be due to specific privacy barriers to a postal delivery HIVST service experienced by this group, outlined in formative work [bib_ref] HIV self-testing among men who have sex with men (MSM) in the..., Witzel [/bib_ref]. Further, our sample reported similar levels of never testing (14%) to other convenience samples of UK MSM (other recent samples range between 8 and 25%), although more participants in the SELPHI pilot had tested in the preceding 12 months [bib_ref] Men who have sex with men in Great Britain: comparing methods and..., Prah [/bib_ref].
When compared to HIVSS return rates in the UK, participants in the pilot made use of their HIVST kits more frequently, lessening missed opportunities for testing. While with HIVSS only 55% of samples are returned for processing [bib_ref] Identifying undiagnosed HIV in men who have sex with men (MSM) by..., Elliot [/bib_ref] , 95% reporting kit use at three months and at least 71% of kits were used overall. At this modest scale, HIVST appears to outperform HIVSS.
Acceptability and ease of use was very high, and indeed higher than in many other studies with MSM [bib_ref] Attitudes and acceptability on HIV self-testing among key populations: a literature review, Figueroa [/bib_ref]. This is not without precedent, with similar levels of acceptability and reported ease of use observed in other settings [bib_ref] Acceptability and feasibility of HIV self-testing among transgender women in San Francisco:..., Lippman [/bib_ref] [bib_ref] Feasibility and acceptability of HIV self-testing among pre-exposure prophylaxis users in Kenya, Ngure [/bib_ref] [bib_ref] Using Grindr™, a smartphone social networking application, to increase HIV self-testing among..., Huang [/bib_ref]. These studies however provided oral fluid HIVSTs, which may have benefits in terms of simplicity (though have lower sensitivity and specificity) over kits which require self-collection of a whole blood sample [bib_ref] Attitudes and acceptability on HIV self-testing among key populations: a literature review, Figueroa [/bib_ref] [bib_ref] Systematic review on HIV self-testing (HIVST) performance and accuracy of results, Figueroa [/bib_ref].
Qualitative accounts of acceptability focused on COM-B domains related to capability more than opportunity or motivation [bib_ref] The behaviour change wheel: a new method for characterising and designing behaviour..., Michie [/bib_ref]. This could signify that when engaging with this novel testing technology, individuals are often doing so with questions about their own skills and capacity. These concerns may decrease with increased experience with HIVST. Indeed, using the lancet was described as difficult for many, although they managed to use it successfully despite this, and all expected this would improve with experience. An additional focus of enquiry for future study is the experience of those who have reactive tests (both confirmed as positive and subsequently confirmed negative) to better understand their experiences, support needs and any potential harms arising.
A number of changes to the trial design were made as a result of this pilot. The attrition between registration and enrolment surveys (24% of participants overall) posed significant recruitment challenges. For the main roll-out of the RCT the language in the email linking the two surveys was made more motivational, specifically highlighting altruism. In addition, all messages used in the roll-out were designed to more clearly emphasise trans eligibility. In response to the increased costs generated by attrition and to take advantage of advertising efficiencies at larger scales, national rather than regional advertising campaigns were prioritised to increase recruitment volumes. Advertisement messages were also altered, with increasing use of motivational elements. In efforts to increase survey completion rates, the number of reminders was increased from 2 to 3, and delivery times were staggered at different times of the day to account for a variety of employment patterns. These changes were supported by a PPI engagement exercise with SELPHI participants.
# Strengths and limitations
This is the first study in Europe to assess the feasibility of recruiting to an online HIVST RCT. A strength of this pilot is that the design perfectly mimics the full trial. Nevertheless, some limitations are noted. Recruited costs per participant should be treated with some caution. For one, the possibility of being randomised to the no baseline HIVST/SoC arm likely made costs per participant significantly higher than delivering free HIVST to all. In addition, recruitment costs in this pilot phase did not show evidence of diminishing returns per participant randomised, meaning that overall cost per participant could become much higher when recruiting larger numbers.
Test kit usability and intervention acceptability were extremely high when compared to other recent studies [bib_ref] Attitudes and acceptability on HIV self-testing among key populations: a literature review, Figueroa [/bib_ref]. A possible explanation is the informed consent procedures in place provided a great deal of information about what a participant could expect from the study and the kit itself in a level of detail that a service might not include.
This pilot struggled to recruit large numbers of trans people compared with cis gender MSM. In addition, sexual practice among MSM is diverse. As only MSM who report lifetime anal sex were eligible for inclusion our sample may not be fully representative of the diversity in MSM sexual behaviour, as between 8 and 19% have never had anal sex. This issue may be an especially pronounced for trans MSM and may have contributed to the low numbers of trans people recruited in this pilot phase.
Finally, while the qualitative data is illuminative, interview were conducted with a small group of participants. The data presented here should be understood as highlighting the diversity of facets of kit usability and intervention acceptability.
[fig] Figure 1: Intervention diagram with COM-B categorisation and targeted domains [/fig]
[fig] Figure 2: Advertising samplesDemographic features of sample including by recruitment source [/fig]
[fig] Figure 3: Map [/fig]
[fig] Figure 4: Intervention acceptability [/fig]
[table] Table 1: Advertising source data [/table]
[table] Table 2: Participant demographics by advert source and overall IQR interquartile range, CAI condomless anal intercourse, HEQ highest educational qualification [/table]
[table] Table 3: Qualitative sub-study sample [/table]
|
Genetic and Epigenetic Regulation of the Brain-Derived Neurotrophic Factor in the Central Nervous System
The BDNF is required for the development and proper function of the central nervous system, where it is involved in a variety of neural and molecular events relevant to cognition, learning, and memory processes. Although only a functional mature protein is synthesized, the human BDNF gene possesses an extensive structural complexity, including the presence of multiple promoters, splicing events, and 3´UTR poly-adenylation sites, resulting in an intricate transcriptional regulation and numerous messengers RNA. Recent data support specific cellular roles of these transcripts. Moreover, a central role of epigenetic modifications on the regulation of BDNF gene transcription is also emerging. The present essay aims to summarize the published information on the matter, emphasizing their possible implications in health and disease or in the treatment of different neurologic and psychiatric disorders.
# Introduction
The Brain-Derived Neurotrophic Factor (BDNF †), purified from the porcine brain in 1982 by Barde et al., is member of the neurotrophins (NTs) family, is involved in the development of the vertebrate nervous system (NS), and regulates synaptic plasticity in the adult brain, influencing the migration of axons and adjusting the number and size of dendrite spines in neurons. It is also involved in neurogenesisand synaptogenesisand facilitates the physiological mechanism of long-term potentiation (LTP) in hippocampus, so it has been associated with processes of learning and memory. In contrast, the immature isoform (proBDNF) has been associated with the activation of neural apoptosisand facilitation of long-term depression in hippocampus.
In addition to its neurotrophic function, BDNF strongly promotes cell survival in various animal models of neurological disorders such as the Amyotrophic Lateral Sclerosis (ALS), Parkinson's and Alzheimer's diseases, and epilepsy. Moreover, the observation that the in vitro application of some antidepressant drugs increases the levels of BDNFsupports a relevant role of this neurotrophin in the patho-physiology of depression. Overall, this leads to consider that failures in the regulation of BDNF synthesis could be related to diverse neurological and psychiatric disorders, as well as sustains the proposal of its possible therapeutic use.
The BDNF possess a structural and functional complexity reflected in 1) the presence of multiple gene promoters; 2) the expression of multiple transcripts, susceptible to alternative splicing events and/or different polyadenylation patterns; 3) the synthesis of diverse precursor isoforms (pre-pro-BDNF), but only a single mature molecule; 4) the activation of two different receptors regulating opposite effects. In summary, all of these features imply the existence of a very selective molecular mechanism that regulates the proper production of BDNF. This review will attempt to summarize the complex BDNF transcription regulation, taking into consideration relevant epigenetic mechanisms. A strong similarity in sequence and gene structure for BDNF among vertebrates is acknowledged; therefore, although some important aspects of the human BDNF gene are still unknown, inferences can be obtained from other species. For instance, the BDNF 5´ exon sequence identity among the human (Homo sapiens), rodent (Rattus norvegicus, Mus musculus), and fish (Dicentrarchus labrax) ranges from 95 percent to 38 percent, with the exon II showing the highest value (≈90%). In the same way, exon I, IV, and VI are quite similar when rodents and humans are compared. Moreover, they share the same alternative 5´ exon splicing mechanism. The white boxes indicate exons, the numbers below show the base pairs that comprise them. The grey spaces point to introns, and the numbers above the base pair that constitute them. The arrows designate the alternative start transcription sites. ASD represents Alternative Splicing donors, while ASA represents Alternative Splicing acceptor sites; the ATG symbols indicate start translation codons. The PA inscription correspond to alternative poly-adenylation sites, and the TAG mark designates the only termination codon for translation within this gene. It is important to point out that the 5' regions of exon I, II, III, IV, V, Vh, VI, VII, and IX correspond to independent promoters that regulate the expression of at least 17 transcripts and that exon I, II, and VI present CpG islands that in this figure are marked by ovals. Finally, the region c of exon IX marked with dark grey corresponds to the codification region of proBDNF.
We will also attempt to discuss the putative implication of these molecular events to health and disease. Other important cellular processes involved in the proper function of the mature protein as the regulation of post-translational modifications or the constitutive or regulated secretion will not be covered here.
## Human bdnf gene structure
The human BDNF gene is located at chromosome 11, region p13-14. The current expert agreement indicates the existence of 11 exons, nine of which contain a specific promoter that regulates its expression. Although the use of alternative promoters is not uncommon (i.e., the presence of two to three promoters has been described in approximately 50 percent of the human genes, the existence of nine different promoter sequences is an exceptional characteristic of the BDNF gene. Moreover, the intron-exon boundaries possess the ar-chetypical GU-AG consensus signals for alternative splicing events.
Additionally, an interesting but frequently ignored feature of the structure of the BDNF gene is the existence of a 200 kb antisense region that includes 10 exons transcribed from a single promoterwith the ability to synthesize a wide variety of anti-BDNF small non-coding RNAs (miRNAs).
As illustrated in, a number of DNA binding sites for distinct transcription factors have been characterized in the different promoters of the rodent BDNF gene. Interestingly, an increase in intracellular calcium (Ca ++ ) levels has been associated with the activation of these binding sites. This is relevant because it has been described that either the activation of glutamate receptors or voltage-gated calcium channels (VGCC) promotes the differential expression of particular BDNF transcripts. Intriguingly, in cultured primary neurons obtained from a transgenic mice to which the human BDNF gene was inserted, the Calcium Response Element (CaRE), the Nuclear Factor Activated T cells Response Element (NFA-TRE), and the Nuclear Factor κβ Response Element (NFκβRE)-activated by an influx of Ca ++ in the rodent BDNF gene after the glutamate receptor stimulationshowed otherwise to be insensitive to the neural depolarization elicited by the activation of L-type VGCC. These results underscore the notion that the involvement of specific intracellular second messenger pathways (activated by intracellular Ca ++ ) dictates the expression of different BDNF transcripts, possibly via the synthesis and binding of different transcription factors to a particular promoter sequence.
Moreover, the preferential activation of certain promoters could have a significant role in the susceptibility of developing certain neurological or psychiatric disorders, as exemplified in reports on depression, bipolar disorder, schizophrenia, epilepsy, and Alzheimer's disease, where changes in the BDNF expression have been described.
## Epigenetic regulation of bdnf
Other molecular events involved in the regulation of BDNF expression are those related to the epigenetic modifications. In a broad sense, epigenetics refers to the way in which chromatin structure is remodeled without affecting the sequence of nucleotides within the DNA. For many years, these processes were only implicated in cellular differentiation and development; however, it is currently recognized that they are also relevant for differentiated cells. A case in point are neurons in which the cell plasticity, necessary for processes of learning and memory, needs to be long-lasting or permanent; therefore, epigenetic mechanisms could help to explain, for example, why neurons do not actively divide.
Experimental results obtained from in vitro and animal models support the role of epigenetics mechanisms on the regulation of BDNF gene expression. For instance, the in silico analysis of the gene sequence shows a number of CpG islands in promoters (p) pI, pII and pIV of the human gene and pI, pII, pIV, pV, pVI, and pIX in the rodent gene. Moreover, the treatment with the DNA methyl inhibitor 5-azacitidyne (5azadC) stimulates the expression of exons I, IV, V, VIII, and IX in C6 rat glioma cells and exons I, IIIand IVin mouse neuroblastoma cells (Neuro2A). On the other hand, addition of the histone deacetylase inhibitor Trichostatin A (TSA) promotes an increased expression of exons III, VII, and VIII in C6 cells, although it did not affect the expression of any transcript in Neuro2A cells. Nonetheless, in the latter example, an increment in the expression of exon I and IV was detected when higher concentrations of TSA were used, an outcome also linked with an increase in Histone (H) 3 and H4 acetylation in the BDNF pI, an epigenetic mark linked with gene expression. Interestingly, in the latter study , the highly methylated status of the 5' proximal region of the BDNF exon I -but not the exon IV -observed under basal conditions disappears after treatment with 5AzadC. As a whole, these results suggest that there must be an extremely specific epigenetic regulation of BDNF expression regarding promoters and cell types.
On the other hand, several molecules participate in the epigenetic regulation of BDNF. One example is the methyl-CpG binding protein 2 (MeCP2), identified by its ability to add methyl groups in the genomeand for recruiting the type I histone deacetylase (HDAC1) Sin3A. It was recently described in cultured mouse cortical neurons that the MeCP2-Sin3A complex generates a long-term inhibition of the BDNF pIV. Moreover, the experimental observation of the phosphorylation and further dissociation of MeCP2 from this locus, either after depolarization of cultured rat neuronsor following the activation of the cyclic adenosin monophosphate (AMPc) pathway in the human neuroblastoma cells, support a relevant role of MeCP2 in the regulation of BDNF gene expression.
Another relevant instance is the growth arrest and DNA damage-inducible 45B protein (Gaad45B) involved in the demethylation of neurogenesis-related genes. This molecule binds to pIX of the rat gene, promoting hippocampal neurogenesis. Additional germane examples with a putative epigenetic effect over this neurotropin are 1) the transcription factor NFκβ, since a di-methylation of lysine 4 residue at H3 histone on the BDNF pI has been detected in a locus containing a potential NFκβ binding site; and 2) the CREB binding protein (CBP), which adds acetyl groups to histones in other gene sequences (like the Major Histocompatibility Complex class II gene). It is known that CREB binds to different BDNF promoters, although it remains to be demonstrated that the complex CREB/CBP can acetylate histones at BDNF specific loci.
Lastly, the anti-BDNF transcripts (miR-NAs) may also have the potential to regulate the expression of this neurotrophin through epigenetic mechanisms. For instance, in vitro studies show that these molecules can diminish both the expression of mRNA and synthesis of the protein through the recruitment of EZH2, a methyl-transferase promoting the three-methylation of lysine 27 residue at H3 on the BDNF locus, an epigenetic mark frequently linked with repression of transcription. Moreover, the genetically engineered over-expression of miRNA 212 in the rat striatum decreases the BDNF protein levels through deactivation of MeCP2. However, no changes were observed in the expression of BDNF after exposure to an enrichment environment in which an increment of mir124, mir132, mir133, and mir14 were also noted.
It is worth to note that epigenetic marks in different promoters of the murine BDNF gene have been described in relation to several "external/environmental" events like consumption of cocaine, stress in the early stages of life, memory related to fear, voluntary exercise, and enriched environment, as well as after psychopharmacological manipulation . Remarkably, in some of these studies, there is a significant correlation among specific histone modification, DNA methylation patterns at pI, pIV, and pVI, and changes in the expression of the correspondent transcripts. Unfortunately, changes in epigenetic marks related to neuropsychiatric disorders has been analyzed only recently in a limited number of studies. Nonetheless, all these messengers share a common coding region included in exon IX that comprises the complete sequence of the proBDNF molecule. Additionally, this exon contains two poly-adenylation sites generating transcripts with either a long or a short 3' UTR. The putative functional relevance of this structural feature is revealed by animal model studies showing that mRNAs with a long 3' UTR are primarily located in dendritic spinesand only translated in response to neuronal activation. In contrast, those with a short 3'UTR are actively translated in the soma to maintain the protein BDNF basal levels. Furthermore, a preferential location of specific transcripts, in particular cellular compartments, has been described. For example, in the rodent visual cortex and hippocampus, the messengers from exons I and IV are predominantly expressed near the neural soma, whereas those generated from exons II and VI are mainly located in distal dendrites. Interestingly, the aforementioned transcriptional compartmentalization appears to have functional consequences, as indicated by recent in vitro experiments in which the over expression or silencing of these specific BDNF 5´splice variants affect in a spatial fashion the dendritic branching and phosphorylation of the BDNF Tyrosine Kinase B (TrKB) receptor. These experimental data also suggest that the different splice variants could represent a spatial code used by neurons to target the effects of BDNF to distinct neural compartments.
Finally, although the BDNF expression along the ontogeny of the brain has been described elsewhere, the analysis of the different transcripts has been barely addressed. For instance, Pattabiraman et al.noted a pattern of differential expression in the rat visual cortex of various splicing forms at different postnatal days (P). In this brain area, exons IV and VI were noticeable at P13 (a developmental early period linked with the start of eye opening), while exons I and II were detected only until P40 (a mature stage of this sensory cortex and where monocular deprivation has no effect). Moreover, an important increase in the expression of exons IV, and in lesser extent, exon VI, is observed, an effect that remains stable throughout adulthood (P90). Interestingly, it was also found that the expression of exons II, IV, and VI were importantly reduced at P40 in the occipital cortex, as a consequence of temporary blocking the electrical activity of retinal ganglion cells, suggesting a link between the rat visual experience and the BDNF gene transcription during the critical period of formation of ocular dominance columns in the Visual Cortex. Unfortunately, an equivalent developmental event in the human visual cortex has not been yet described.
On the other hand, the developmental BDNF expression in the human Dorsolateral Prefrontal Cortex was recently reported. In this cortical region, exons I and VI were steadily expressed throughout the first phases of infancy (neonate, infant, toddler) but decline in the following years (6 to 40 years). In contrast, transcripts II and IV, barely detected in neonates, increased their expression in infants and toddlers to finally diminish and reach a steady level from 5 years through adolescence and adulthood. These data show important brain regional and temporal variation in BDNF expression and important differences among species.
Another putative implication of the expression of this neurotrophin during development was recently discussed by Calabrese et al., who reported that compared to wild type animals, SERT (serotonin transporter) knockout rats displayed a reduced expression of BDNF transcrits I, IV, VI in ventral hippocampus and prefrontal cortex 1 to 4 weeks after birth. The simultaneous decrease of both SERT and BDNF in these animals seems to explain the evident signs of depression and anxiety observed in these genetically modified animals, probably affecting neural plasticity. Moreover, it is tempting to assume that changes at critical time windows of development of the expression of this neurotrophin can increase the risk for mood/anxiety, particularly in those individual carrying certain genetic variants of the SERT.
In summary, there is solid experimental evidence of the existence of multiple BDNF transcripts with apparently distinct functional properties. Further studies are encouraged, as the accurate time and spatial description of the splice variants might provide key information about the particular cell types or neuronal circuits involved in specific neuropsychiatric disorders, also stimulating the development of alternative pharmacological treatments. Regrettably, only few studies have determined the expression of more than a single transcript at once.
## Post-translational modifications and bdnf isoforms
As illustrated in , exons I, VII, VIII, and IX possess an alternative translation start codon, but only exon IX includes a translation stop codon. Therefore, theoretically, four different pre-pro-BDNF protein isoforms could be synthesized differing in an extended amino terminal region according to the particular transcribed exon (exon I, VI, and VIII: 8, 15 and 81 amino acids, respectively). It has been proposed that the length of the pre-domain could affect the intracellular BDNF trafficking, with the longer versions preferentially promoting the secretion of the immature isoform.
In the brain, the 32 kDa proBDNF can experience at least three final paths: 1) to be edited mainly in Golgi and secreted as the mature BDNF molecule; 2) to be secreted as proBDNF and procesed to mature BDNF in the synaptic space; or 3) to be secreted as proBDNF without any subsequent digestion. In any case, the proper understanding of the different mechanisms and circumstances associated with the neurotrophin secretion will be of great biological relevance as the fine-tuned equilibrium between the different isoforms could dictate different physiological outputs.
ProBDNF binds to the receptor p75 (a member of the tumor necrosis factor receptor family). This binding has been associated with the boosting of long-term depression in neurons of the hippocampusand neuronal pruning in the nervous systemor with initiating programmed neuronal death. Similar effects have also been reported when proBDNF binds to the hetero-dimeric p75/sortilin receptor. Moreover, the p75 receptor can also be associated with the Nogo receptor complex (NogoR, Lingo, p75NTR) that mediates inhibition of axon growth. In any case, the binding of proBDNF to p75 activates the Jun kinase signal cascade pathway (JNK), more specifically triggering JNK3, causing apoptosis through the activation of p53 tumor suppressor gene and caspases. Moreover, proBDNF/p75 also activates the small GTPase RhoA and its downstream effector Rho Kinase that has been associated to inhibition of neurite outgrowth.
On the other hand, the mature BDNF molecule has an important affinity for TrKB (Kd = 9.9 × 10 −10 M). Binding to this membrane receptor has been related to the increase of synaptic transmission and plasticity, neural proliferation and survival, and axonal sprouting. In the absence of TrKB, mature BDNF could also bind to p75 receptor in order to regulate axon pruning. The BDNF/TrKB interaction leads to the activation of three intracellular signaling cascades: 1) the phospholipase C; 2) the phosphatidyl-inositol-3 phosphate kinase (PI3K); or 3) the kinases regulated by extracellular signals. These mechanisms promote calcium entrance into the cell that leads to an increment in gene expression, associated with cellular differentiationand dendrite formation. Moreover, TrKB activation stimulates AKT signaling pathway through PI3K to modulate cell survival.
As previously mentioned, the transcripts with a long 3'UTR are located mainly to distal dendrite. This could largely lead to the secretion of the proBDNF molecule, as this neural compartment does not possess Golgi. Remarkably, in the fish Dicentrarchus labrax, the differential expression of BDNF transcripts after stress can produce an increment in the secretion of proBDNF. To our knowledge, no single study has evaluated if the differential expression of BDNF transcripts could affect the proportion of which proBDNF and BDNF isoforms are secreted in the human brain, information that could have important implications for neuropsychiatric disorders.
# Conclusion
As revealed by the increasing number and variety of papers in recent years related to a very broad spectrum of BNDF-associated topics, including health, disease, neuroscience, or cognition, this neurotrophin has become a paradigmatic example of a "multitask" neural molecule.
More specifically, the multiple neural functions associated with its effects in the nervous system have foreseen certain therapeutic applications for a variety of neurological and psychiatric disorders, including ALS, epilepsy, depression, Parkinson's, and Alzheimer's disease. Of special interest is the case of ALS, where the administration of BDNF (and other neurotrophins) has been already evaluated in humans. Although the preclinical studies appeared promising, phase III trials showed only minimal beneficial effects for a subgroup patients (e.g., those in an advanced stage of the disease), a failure probably related to the complex regulation of BDNF within the nervous system. Moreover, the recent description of an accurate distinction between patients with major depression or healthy controls, based on the methylation profiles of CpG units within the BDNF pI in peripheral samples (e.g., leucocytes, blood mononuclear cells), envisage its putative use as an efficient diagnostic biomarker of depression. This approach has been recently attempted for other psychiatric conditions such as schizophrenia, borderline personality disorder (BPD), depression after a stroke event, and bipolar disorder, or as a surrogate marker of clinical response for psychopharmacological response in depressionor psychotherapy in BPD. Interestingly, in the latter example, the methylation status of BDNF was positively associated with the history of child maltreatment, showing that this neurotrophin is an important modulator of the gene/environment interplay, as we have previously reported by evaluating the genetic variance of BDNF in depressive patients.
As it has been highlighted in this review, the functional effects of this molecule in a particular brain region would depend on multiple time/location-regulated molecular events, including the synthesis of their different transcripts; the presence of specific transcription factors; the recruitment of molecules with epigenetic effects; and the activation of particular receptors; among others. Moreover, epigenetic mechanisms that seem to be strongly dependent on external/environmental events must be included in this complex molecular equation. This is particularly important to neuropsychiatry stimulating new avenues for clinical experimentation (e.g., by using methylation/acetylation histones-modifying drugs) as currently attempted in cancer treatment.
In any case, further investigation at clinical and basic levels of BDNF production is warranted. |
Toxic shock syndrome and pyomyositis: about an unusual case
Pyomyositis is a pyogenic infection of skeletal muscle with abscess formation. It is a rare disease with nonspecific symptoms which requires a rapid diagnosis and treatment. Magnetic resonance imaging is considered the gold standard for early diagnosis and to rule out other etiologies. This article reports an atypical presentation of pyomyositis revealed by a toxic staphylococcal shock syndrome in an 8-year-old boy.
# Introduction
Pyomyositis is a primary infection of the skeletal muscle usually with abscess formation. It is a rare condition initially described in 1885, occurring classically in tropical countries. Its pathogenesis is not yet fully understood, however, it is thought to be induced by bacteremia where staphylococcus is the most accused germ. Throughout this study, we report the case of pyomyositis revealed by a toxic staphylococcal shock syndrome. calcaneus on short-TI inversion recovery (STIR) images suggesting bilateral pyomyositis. Indeed, the delay between first signs and diagnosis was 2 months. Intravenous antibiotic therapy was started with amoxicillin-clavulanic acid for 4 weeks then 15 days orally, for a total duration of 6 weeks, associated with motor physiotherapy. The patient had after 20 days of evolution a drainage of abscesses by the team of orthopedic surgeons of our university hospital, however, the puncture was white; complemented by an incision which did not objectify pus with a sterile bacteriological study. Fever disappeared after 72 hours; the child gradually resumed walking during his hospitalization with a complete regression of the symptomatology after 6 weeks. Regarding the search for an underlying cause, the HIV serology was negative, the fasting blood sugar was 0.92g/dL, the lactate -ammonia level were normal and the dihydrorhodamine 123 test (DHR) for a chronic septic granulomatosis was normal. However, the assessment of immunoglobulins, lymphocyte subpopulations and vaccine antibodies was not carried out.
## Patient and observation
# Discussion
Pyomyositis or tropical myositis is a suppurative disease of the striated muscle originally described by Scriba in Japan. The clinical symptomatology is insidious and not very specific, associating a fever, pain with inflammation of the involved location as well as general signs of sepsis. In rare cases the presentation is that of a toxic shock syndrome, this was the case in our patient. Indeed, it is not a very well-known disease simulating several pathologies which explains the diagnostic delay with an average delay described in the literature of 15 days (10-62 days) joining our study, hence the interest of recognizing pyomyositis in children in front of any painful and febrile lameness. Several authors have reported the concept of limb trauma as one of the factors favoring abscess formation with a predilection for the muscles of the lower limb and the small pelvis: psoas, external and internal obturators, quadriceps and adductor, which is also the case for our patient. The increase in biological parameters (neutrophil leukocytosis, CRP) is the same regardless the topography, blood cultures are crucial for the identification of a germ but are positive in only half of the casesaccordingly the interest of bacteriological sampling of abscesses.
In our case, the two blood cultures were positive with isolation of Staphylococcus aureus (SA) which remains the most involved germ in the literature with a rate of 60%, but other germs can be responsible for pyomyositis such as streptococcus or anaerobes.
The role of radiological assessment in the diagnosis of pyomyositis is essential. X-rays allow eliminating certain differential diagnoses such Page number not for citation purposes 3 as osteomyelitis or arthritis, ultrasound is sensitive and keeps its place in the drainage of abscesses and MRI remains the radiological examination of choice given its sensitivity and specificity; it allows the diagnosis to be made at an earlier stage. Concerning the treatment of pyomyositis, it is not codified; the empirical antibiotic therapy of choice in the literature remains penicillin M intravenously for ten days since the main germ is the SA sensitive to methicillin with oral relay adapted to the antibiogram for a total duration of 4 to 6 weeks on average, for our patient we prescribed amoxicillin-clavulanic acidconsidering the necrosis complication encountered with penicillin M during many experiences in our department and also that amoxicillin-clavulanic acid is active on staphylococcus and provides a good clinical improvement.
# Conclusion
Pyomyositis is an unrecognized and nonspecific disease that should always be kept in mind in front of clinical triad of fever, pain and lameness. Blood cultures and abscess samples can most often isolate a germ, but MRI remains the benchmark for positive diagnosis. |
Specific Gene Repression by CRISPRi System Transferred through Bacterial Conjugation
The E. coli strain S17-1 was purchased from ATCC, and the recipient E. coli strain, an MJ1655 strain with chromosomal insertion of sfGFP and mRFP, was obtained from Lei S. Qi 1 . E. coli strain Top10 (Life Technologies) was used for cloning purposes.Plasmid pARO190 was purchased from ATCC. Two additional plasmids, pdCas9_bacteria containing S. pyogenes dCas9 protein-coding gene under control of an anhydrotetracycline (aTc)-inducible promoter (pLTetO-1), and pgRNA_RFP containing sgRNA placed under a constitutive promoter (iGEM Parts Registry BBa_J23119), were obtained from Lei S. Qi 1 .E. coli strains transformed with plasmids pARO190, pdCas9_bacteria, and pgRNA_RFP, were maintained in 100 ug/mL carbenicillin, 34 ug/mL chloramphenicol, and 100 ug/mL carbenicillin, respectively. Recipient E. coli strain was routinely maintained in 50 g/mL kanamycin, and strain S17-1 was routinely maintained in 100 g/mL spectinomycin. All E. coli strains were cultured in LB (Luria-Bertani) broth in culture tubes at 37 °C with 300 rpm shaking speed unless otherwise specified.
## Supplemental
# Supplementary methods
## Plasmid construction
To construct plasmid dCas9-pARO190, DNA fragments of dCas9 and pARO190 backbone were amplified by PCR from plasmids pdCas9_bacteria and pARO190, respectively, with a 20 bp overhang for each fragment. Primers 1 and 2 were used to amplify dCas9, and primers 7 and 8 were used to amplify pARO190 backbone. GeneArt® Seamless Cloning and Assembly Enzyme Mix (Life Technologies) was then used for assembly of the two fragments.
To construct plasmid dCas9-sgRNA-pARO190, DNA fragments of dCas9, sgRNA and pARO190 backbone were amplified by PCR from plasmids pdCas9_bacteria, pgRNA_RFP, and pARO190, respectively, with a 20 bp overhang for each fragment. Primers 1 and 3 were used to amplify dCas9, 4 and 5 for sgRNA, and 6 and 8 for pARO190 backbone. GeneArt® Seamless Cloning and Assembly Enzyme Mix (Life Technologies) was then used for assembly of the three fragments.
Primers were chemically synthesized at Integrated DNA Technologies Inc, and PCR amplification was performed with Phusion High-Fidelity DNA Polymerase (New England Biolabs). Chemical transformation was used for cloning purposes, and electroporation was used to transform plasmids into E. coli strain S17-1.
## Conjugation
Overnight cultures of donor and recipient strains were washed three times by centrifuging at 3000 rpm for 2 minutes and then resuspending in fresh LB broth without antibiotics. The resuspended cultures were diluted and co-cultured into 10 mL fresh LB broth without antibiotics, to OD 600 of 0.05 for both strains in the co-culture. The co-culture was then incubated at 37°C with 100 rpm shaking speed to allow for conjugation. After a specific time, the co-culture was vigorously mixed on a vortex mixer to stop conjugation and separate connecting donor and recipient cells.
## Calculation of conjugation efficiency
S17-1 strains transformed with dCas9-sgRNA-pARO190 served as donor strains and the reporter strains served as recipients. After conjugation, co-cultures were plated on LB agar plates containing antibiotics of spectinomycin, kanamycin, and both carbenicillin and kanamycin, for the selection of donor, recipient, and trans-conjugant strains respectively. Serial dilutions (10 3 , 10 4 , 10 5 , 10 6 fold) of the co-cultures were made prior to plating, and 200 l of each dilutions were plated. After overnight incubation, colonies on each plate were counted, and the results were presented as colony forming units per milliliter culture (cfu/ml). Conjugation efficiencies were calculated by dividing counting results of trans-conjugant cells to that of recipient cells. Time points of 2, 5, and 8 hours of conjugation were tested for estimation of optimal conditions.
## Supplemental table 2: conjugation efficiency
## Induction and flow cytometry
The reporter strains were separately co-cultured with S17-1 strains transformed with plasmids dCas9-pARO190 and sgRNA-dCas9-pARO190 (conjugative plasmid) for 8 hours for conjugation. Co-cultures were then diluted 10 4 fold and plated on LB agar plates containing both carbenicillin and chloramphenicol to select for trans-conjugants. After overnight incubation, liquid cultures of trans-conjugant colonies were made and grown overnight for saturation. The overnight cultures were then diluted to OD 0.05 into fresh carbenicillin and chloramphenicol containing LB broth in 2 ml 96-well deep well plates in duplicate, with 10 g/mL aTc supplemented to one replicate of each conjugation to induce production of the dCas9 protein.
Cultures were incubated at 37°C with shaking (1200 rpm) for 8 hours, washed and resuspended in PBS, and run on a LSRII flow cytometer (BD Biosciences) equipped with a high-throughput sampler to determine the levels of fluorescent proteins. Recipient strains without conjugation
[table] Table 1: Primer Sequences [/table]
|
Noninvasive intracranial pressure estimation by orbital subarachnoid space measurement: the Beijing Intracranial and Intraocular Pressure (iCOP) study
Introduction:The orbital subarachnoid space surrounding the optic nerve is continuous with the circulation system for cerebrospinal fluid (CSF) and can be visualized by using magnetic resonance imaging (MRI). We hypothesized that the orbital subarachnoid space width (OSASW) is correlated with and can serve as a surrogate for intracranial pressure (ICP). Our aim was to develop a method for a noninvasive measurement of the intracranial CSF-pressure (CSF-P) based on MRI-assisted OSASW.Methods: The prospective observational comparative study included neurology patients who underwent lumbar CSF-P measurement and 3.0-Tesla orbital magnetic resonance imaging (MRI) for other clinical reasons. The width of the orbital subarachnoid space (OSASW) around the optic nerve was measured with MRI at 3, 9, and 15 mm behind the globe. The study population was randomly divided into a training group and a test group. After adjusting for body mass index (BMI) and mean arterial blood pressure (MABP), algorithms for the associations between CSF-P and OSASW were calculated in the training group. The algorithms were subsequently verified in the test group. Main outcome measures were the width of the orbital subarachnoid space (OSASW) and the lumbar cerebrospinal fluid pressure (CSF-P). Results: Seventy-two patients were included in the study. In the training group, the algorithms for the associations between CSF-P and OSASW were as follows: (a) CSF-P = 9.31 × OSASW (at 3 mm) + 0.48 × BMI + 0.14 × MABP-19.94; (b) CSF-P = 16.95 × OSASW (at 9 mm) + 0.39 × BMI + 0.14 × MABP-20.90; and (c) CSF-P = 17.54 × OSASW (at 15 mm) + 0.47 × BMI + 0.13 × MABP-21.52. Applying these algorithms in the independent test group, the measured lumbar CSF-P (13.6 ± 5.1 mm Hg) did not differ significantly from the calculated MRI-derived CSF-P (OSASW at 3 mm: 12.7 ± 4.2 mm Hg (P = 0.07); at 9 mm: 13.4 ± 5.1 mm Hg (P = 0.35); and at 15 mm: 14.0 ± 4.9 mm Hg (P = 0.87)). Intraclass correlation coefficients (ICCs) were higher for the CSF-P assessment based on OSASW at 9 mm and at 15 mm behind the globe (all ICCs, 0.87) than for OSASW measurements at 3 mm (ICC, 0.80).Conclusions: In patients with normal, moderately decreased or elevated ICP, MRI-assisted measurement of the OSASW appears to be useful for the noninvasive quantitative estimation of ICP, if BMI and MABP as contributing parameters are taken into account.
# Introduction
Knowledge of intracranial pressure (ICP) is of major importance for the diagnosis of neurologic and neuroophthalmologic diseases. The ICP has been measured invasively by lumbar puncture [bib_ref] CSF pressure assessed by lumbar puncture agrees with intracranial pressure, Lenfeldt [/bib_ref]. Noninvasive methods that were explored to estimate the ICP included transcranial Doppler sonography [bib_ref] Clinical assessment of noninvasive intracranial pressure absolute value measurement method, Ragauskas [/bib_ref] , tympanic membrane displacement measurement [bib_ref] Mean intracranial pressure monitoring by a non-invasive audiological technique: a pilot study, Reid [/bib_ref] , computed tomography [bib_ref] Computed tomography in benign intracranial hypertension, Weisberg [/bib_ref] , magnetic resonance imaging (MRI) [bib_ref] MR-Intracranial pressure (ICP): a method to measure intracranial elastance and pressure noninvasively..., Alperin [/bib_ref] , scanning laser tomography of the optic nerve head [bib_ref] Laser scanning tomography of the optic nerve vs CSF opening pressure in..., Heckmann [/bib_ref] , and venous ophthalmodynamometry [bib_ref] Ophthalmodynamometry: a reliable method for measuring intracranial pressure, Motschmann [/bib_ref]. All these techniques, however, had some limitations, such as that transcranial Doppler sonography cannot be used on 10% to 15% of the patients because of the ultrasound not being able to penetrate the skull [bib_ref] Advances in transcranial Doppler ultrasonography, Tsivgoulis [/bib_ref] ; venous ophthalmodynamometry could be applied only in patients with elevated ICP without papilledema [bib_ref] Ophthalmodynamometric measurement of central retinal vein pressure as surrogate of intracranial pressure..., Jonas [/bib_ref] ; or because of the perilymphatic duct being less passable with age, tympanic membrane displacement measurements have a relatively low practicability.
The orbital subarachnoid space around the optic nerve is continuous with the cranial subarachnoid space via the optic nerve canal and can be visualized by using T 2weighted MRI with a fat-suppressed sequence [bib_ref] Fast and quantitative high-resolution magnetic resonance imaging of the optic nerve at..., Weigel [/bib_ref]. The pressure in the orbital subarachnoid space is correlated with the ICP [bib_ref] Measurement and relationship of subarachnoid pressure of the optic nerve to intracranial..., Liu [/bib_ref]. Patients with increased ICP have a wider than normal orbital subarachnoid space and vice versa [bib_ref] Ocular sonography in patients with raised intracranial pressure: the papilloedema revisited, Geeraerts [/bib_ref] [bib_ref] Optic nerve sonography in the diagnostic evaluation of adult brain injury, Soldatos [/bib_ref] [bib_ref] Using MRI of the optic nerve sheath to detect elevated intracranial pressure, Kimberly [/bib_ref] [bib_ref] Effect of intracranial pressure on the diameter of the optic nerve sheath, Watanabe [/bib_ref] [bib_ref] Decreased diameter of the optic nerve sheath associated with CSF hypovolemia, Watanabe [/bib_ref] [bib_ref] MR imaging of the optic nerve sheath in patients with craniospinal hypotension, Rohr [/bib_ref] [bib_ref] Highresolution magnetic resonance imaging of the intraorbital optic nerve and subarachnoid space..., Mashima [/bib_ref]. These findings led to the hypothesis that the orbital subarachnoid space width (OSASW) is correlated with, and could serve as a surrogate for, the ICP. Further support for this hypothesis was provided by a study reporting a linear relation between the optic nerve sheath diameter, as measured by sonography, and the lumbar cerebrospinal fluid pressure in 12 patients [bib_ref] Validation of the optic nerve sheath response to changing cerebrospinal fluid pressure:..., Hansen [/bib_ref]. In a similar manner, the optic nerve sheath diameter, as measured with MRI, significantly correlated with the ICP in patients with traumatic brain injury [bib_ref] Use of T2-weighted magnetic resonance imaging of the optic nerve sheath to..., Geeraerts [/bib_ref]. These studies, however, had limitations, such as using sonography with a relative precision for measurements of the diameter of the optic nerve and the OSASW [bib_ref] Morphometry of the retrobulbar human optic nerve: comparison between conventional sonography and..., Lagrèze [/bib_ref] , or the studies did not quantitatively assess the ICP [bib_ref] Ocular sonography in patients with raised intracranial pressure: the papilloedema revisited, Geeraerts [/bib_ref] [bib_ref] Optic nerve sonography in the diagnostic evaluation of adult brain injury, Soldatos [/bib_ref] [bib_ref] Using MRI of the optic nerve sheath to detect elevated intracranial pressure, Kimberly [/bib_ref] [bib_ref] Effect of intracranial pressure on the diameter of the optic nerve sheath, Watanabe [/bib_ref] [bib_ref] Decreased diameter of the optic nerve sheath associated with CSF hypovolemia, Watanabe [/bib_ref] [bib_ref] MR imaging of the optic nerve sheath in patients with craniospinal hypotension, Rohr [/bib_ref] [bib_ref] Highresolution magnetic resonance imaging of the intraorbital optic nerve and subarachnoid space..., Mashima [/bib_ref] [bib_ref] Validation of the optic nerve sheath response to changing cerebrospinal fluid pressure:..., Hansen [/bib_ref] [bib_ref] Use of T2-weighted magnetic resonance imaging of the optic nerve sheath to..., Geeraerts [/bib_ref] , or the studies addressed only special clinical situations such as acute brain trauma, or the diameter of the optic nerve sheaths as surrogate for the OSASW was measured without taking into account the diameter of the optic nerve.
To avoid these limitations, we conducted this study to test the hypothesis whether the OSASW, as measured by orbital MRI, can be used to estimate the ICP.
# Material and methods
The prospective observational comparative study included patients who consecutively underwent cranial MRI and a lumbar puncture for diagnosis and treatment of neurologic diseases between June 2011 and April 2012. The study protocol was approved by the Medical Ethics Committee of the Beijing Tongren Hospital, according to the Declaration of Helsinki, and all patients signed a written informed consent. The study was registered in the Chinese Clinical Trial Registry (registration site: ChiCTR-OCC-11001271). Exclusion criteria for the study were bilateral optic neuritis, optic nerve tumors, ocular or intracranial tumors, visual acuity worse than 20/400, any orbital disease, any cranial surgery, traumatic brain injury, previous lumbar puncture, which may cause hemorrhage within the CSF circulation system and result in obstruction of the spinal subarachnoid space, and the inability to perform an MRI examination properly.
All patients underwent a complete neurologic and ophthalmologic examination, cranial and orbital MRI, and lumbar CSF-P measurement. Body weight and height were measured. The ophthalmologic examination included visual acuity assessment, refractometry, tonometry, slit lamp-assisted biomicroscopy of the anterior and posterior segment of the eye, ophthalmoscopy, and peripapillary retinal nerve fiber layer thickness measurement with spectral domain optical coherence tomography (RTVue-100; software version 4.0; Optovue, Inc., Fremont, CA, USA).
The MRI of the orbital part of the optic nerve/sheath complex was performed at 14:00 hours in a standardized manner in supine position. We used a 3.0-Tesla wholebody scanner (Signa HDx; General Electric Medical System, Milwaukee, WI, USA) equipped with an eightchannel phased-array head coil. To avoid artifacts due to motion of the eye, all subjects were instructed to fixate on a target attached directly in the gantry of the MRI scanner with the eye in primary gaze. Both eyes of the patients were examined in the same manner. If a motion artifact was detected during the study, the sequence was repeated.
For the measurement of the optic nerve/sheath complex, a fast-recovery fast spin-echo sequence (FRFSE) was applied, as described in detail previously [bib_ref] Orbital cerebrospinal fluid space in glaucoma: the Beijing Intracranial and Intraocular Pressure..., Wang [/bib_ref]. Scout images in the transverse and oblique sagittal planes were used to ensure optimal head positioning; oblique coronal images were used for quantification. Two basic FRFSE sequences were used: [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref]. -A T 2 -weighted fast-recovery fast spin-echo sequence with fat suppression was optimized for quantification of the morphology (TR = 6,000 milliseconds; TE = 245 milliseconds; number of excitations = 2; echo-train length = 60; bandwidth = 20.83 Hz/pixel; field of view =16 cm × 16 cm; matrix = 320 × 320; nominal spatial resolution = 0.5 mm × 0.5 mm; slice thickness = 3 mm; slice gap = 0).
[formula] -A T 2 -weighted [/formula]
The T2WI-FRFSE images were interpolated to a higher matrix size of 1,024 × 1,024, leading to a pixel size of 0.16 mm × 0.16 mm for better visualization. Seven oblique coronal MR images were continuously acquired perpendicular to the optic nerve orientation with placement of the first slice directly posterior to the globe. The images were acquired for both eyes separately [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref]. The optic-nerve acquisition time was 1′18″. The acquisition time was 11 seconds per slice. In these oblique coronal images, the cerebrospinal fluid (CSF) showed a high, white signal, and the optic nerve parenchyma, a low, dark signal [fig_ref] Figure 2: Scheme to demonstrate the optic nerve/sheath complex [/fig_ref]. Three oblique coronal slices perpendicular to the optic nerve at 3, 9, and 15 mm behind the globe were evaluated. Two experienced radiologists (YL, WC) evaluated the images in a masked manner, by using the postprocessing Advantage Workstation 4.4 software (General Electric, Milwaukee, WI, USA).
The horizontal and vertical diameters of the optic nerve and the optic nerve sheath were measured. The average diameter of the optic nerve and of the optic nerve sheaths was calculated as the mean of the measured horizontal and vertical diameters. The width of the optic nerve subarachnoid space was calculated as the difference of half of the optic nerve sheath diameter minus half of the optic nerve diameter [fig_ref] Figure 2: Scheme to demonstrate the optic nerve/sheath complex [/fig_ref] [bib_ref] Subarachnoid fluid of the optic nerve in normal adults, Lam [/bib_ref]. The measurement results of the first observer were used for the primary analysis. The inter-and intraobserver repeatability was tested on 30 randomly selected individuals from all the patients. For the assessment of the intraobserver repeatability, observer 1 performed the same analysis twice at an interval of 3 months.
The lumbar CSF-P was measured by the same neurologist (GT) in a standardized manner at 14:00 hours in a lateral decubitus position, with the patient's neck bent in full flexion and the knees bent in full flexion up to the chest. A standard spinal needle (20-gauge, 90 mm in length) was used. The opening pressure was measured. During the procedure, all patients were awake and not sedated. Systolic and diastolic blood pressure was measured in the supine position just before the lumbar puncture was performed. Mean arterial blood pressure was calculated as 1/3 × systolic blood pressure + 2/3 × diastolic blood pressure.
Statistical analysis was performed by using a commercially available statistical software package (SPSS for . The study population was randomly assigned to a training group and a test group, in a ratio of 4:3. Only one randomly selected unaffected eye per patient was taken for statistical analysis. We determined the mean value (presented as mean ± standard deviation) of the main outcome parameters. The distribution of the values was assessed by using the Kolmogorov-Smirnov test. Differences in the demographic, ophthalmologic, and intracranial characteristics between the training group and the test group were then assessed by using two-tailed Student t test. Proportions were compared by using the χ 2 test. All P values were two-sided.
In a first step of the statistical analysis, we used the data of the training group and performed a univariate analysis of the associations between the lumbar CSF-P, MRI-derived orbital measurements, body mass index, mean arterial blood pressure, age, intraocular pressure, and retinal nerve fiber layer thickness.
In a second step, linear, quadratic, and cubic regression models in a multivariate analysis were constructed to assess the associations between lumbar CSF-P and those parameters, which were significantly associated with CSF-P in univariate analysis. Good fitness values and parsimony of the three regression models were compared. The best fit (judged by the r 2 value) and the most parsimonious one was chosen. Durbin-Watson statistic was used to test for the presence of serial correlations among the residuals [bib_ref] Analysis of residuals: criteria for determining goodness-of-fit, Straume [/bib_ref]. A test for collinearity was performed to test for possible multicollinearity among the independent parameters. A Durbin-Watson statistic between 1.5 and 2.5 indicated that no serious residual autocorrelation was present.
In a third step of the analysis, we tested the value of the calculated mathematical functions to predict the ICP in the test group. A Bland-Altman analysis was applied to measure the prediction's accuracy and precision [bib_ref] Statistical methods for assessing agreement between two methods of clinical measurement, Bland [/bib_ref]. The intraclass correlation coefficient (ICC) and 95% confidence intervals (CIs) of the comparison of both methods were calculated to determine the prediction's reliability. These procedures were also used to assess the interand intraobserver repeatability of the morphologic MRI evaluations.
# Results
The study included 72 Han Chinese patients (mean age, 42.0 ± 13.4 years; range, 19 to 70 years), with the data of 42 patients assigned to the training group and the data of the other 30 patients assigned to the test group. The indications for lumbar puncture were peripheral neuropathy, intracranial hypertension, spontaneous intracranial hypotension, cavernous sinus syndrome, meningitis, multiple sclerosis, unilateral ischemic optic neuropathy, unilateral optic neuritis, optic nerve atrophy, and head injury. Because of randomization, the training group and test group did not differ significantly in age, gender, body height and weight, body mass index, intraocular pressure, retinal nerve fiber layer thickness, and arterial blood pressure (all P > 0.10). The MRI scans of the OSASW taken at 3 mm behind the globe could be assessed for all patients. Because of image-quality problems, the MRI scans of the OSASW taken at 9 mm behind the globe could not be assessed for three (4.1%) patients, and the MRI scans taken at 15 mm behind the globe could not be assessed for seven (9.5%) patients. Patients with elevated ICP have a wider orbital subarachnoid space than do the patients with decreased ICP [fig_ref] Figure 3: See legend on next page [/fig_ref].
Including all study participants, the mean optic nerve diameter at 3, 9, and 15 mm behind the globe was 3. [bib_ref] Effect of intracranial pressure on the diameter of the optic nerve sheath, Watanabe [/bib_ref] In the training group, lumbar CSF-P was strongly correlated with the OSASW at 3, 9, and 15 mm behind the globe (Pearson correlation r: 0.83 ≤ r ≤ 0.88; all P < 0.0001) [fig_ref] Figure 4: Scattergram showing the distribution of lumbar cerebrospinal fluid pressure measurements versus the... [/fig_ref]. The correlation coefficients for these correlations were higher than those for the associations between lumbar CSF-P and the optic nerve sheath diameters (0.66 ≤ r ≤ 0.76; all P < 0.0001). The retinal nerve fiber layer thickness was not significantly associated with the OSASW measured at 3, 9, and 15 mm behind the globe (P = 0.15; P = 0.66; and P =0.34, respectively). It was significantly associated with the optic nerve diameter at 9 mm (r = 0.61; P = 0.001) and 15 mm (r = 0.75; P < 0.0001) and with the optic nerve sheath diameter at 9 mm (r = 0.57; P = 0.003) and at 15 mm (r = 0.75; P < 0.0001). The lumbar CSF-P values were additionally significantly associated with the body mass index (r = 0.61; P < 0.0001) and mean arterial blood pressure (r = 0.55; P < 0.0001). In our study population, lumbar CSF-P was not significantly associated with age (P = 0.22), intraocular pressure (P = 0.67), or retinal nerve fiber-layer thickness (P = 0.47).
In the second step of the statistical analysis, different models were tested for the associations between the OSASW and lumbar CSF-P values. The r 2 was roughly equal in the linear, quadratic, and cubic models, with the linear model being the most parsimonious. Strong positive linear relations between lumbar CSF-P measurements and the OSASW at 3, 9, and 15 mm were determined within a CSF-P range from 3.7 to 26.5 mm Hg. Similarly, lumbar CSF-P showed a moderately positive linear relation with body mass index (BMI) and mean arterial blood pressure (MABP).
In the third step, a stepwise multivariate linearregression analysis revealed that the OSASW (at 3, 9, and 15 mm behind the globe), body mass index, and mean arterial blood pressure were independently associated with lumbar CSF-P measurements. They were entered into multiple regression models , out of which three weighting functions for the prediction of ICP derived: In the fourth step of the statistical analysis, the three algorithms derived earlier were applied in the test group. It showed that the measured lumbar CSF-P (13.6 ± 5.1 mm Hg) did not differ significantly from the calculated MRI-derived mean ICP values obtained by using the three weighting functions (weighting function 1, 12.7 ± 4.2 mm Hg (P = 0.07); weighting function 2, 13.4 ± 5.1 mm Hg (P = 0.35); and weighting function 3, 14.0 ± 4.9 mm Hg (P = 0.87). The mean estimated bias ± standard deviation and the 95% limits of agreement suggested that greater accuracy and precision was achieved from functions (2) and (3) than from function (1). The results of the Bland-Altman analysis are shown in [fig_ref] Figure 5: Bland-Altman graph of the inter-method agreement of the non-invasive magnetic resonance imaging [/fig_ref]. All intraclass correlation coefficients (ICCs) of the two methods reached values ≥0.80. The ICCs and their 95% CIs for the noninvasive ICP assessment by using function (1), (2), and (3) were 0.80 (0.62 to 0.90), 0.87 (0.74 to 0.94), and 0.87 (0.74 to 0.94), respectively, which suggested that the reliability of the noninvasive ICP assessment using functions (2) and (3) was higher than the function [bib_ref] CSF pressure assessed by lumbar puncture agrees with intracranial pressure, Lenfeldt [/bib_ref].
The data on the intraobserver and interobserver reproducibility of the optic nerve-sheath complex measurements by MRI, including the 95% limits of agreement of the mean differences, suggested that the intraobserver agreement was superior to the interobserver agreement [fig_ref] Table 3: Interobserver and intraobserver repeatability of the optic nerve-sheath complex measurements by MRI... [/fig_ref]. For both, intraobserver measurements and interobserver measurements, the ICCs, and their 95% CIs were better for the measurements taken closer to the globe.
# Discussion
Within the range of a lumbar CSF-P between 3.7 mm Hg and 26.5 mm Hg, the ICP determined by lumbar puncture was significantly correlated with the OSASW, as determined with MRI. Good fitness was roughly equal in the linear, quadratic, and cubic models of the relation between MRI-determined OSASW and lumbar CSF-P. Because the linear model was the most parsimonious, we chose the linear-fitting equation for further analysis. After adjusting for body mass index and mean arterial blood pressure, two formulas were formed to calculate the ICP, based on the MRI measurements of the OSASW at 9 mm or at 15 mm behind the globe. Applied in an independent test group, these formulas were relatively precise in predicting the ICP. The 3.0-T MRI protocols, based on T2WI-FRFSE with fat-suppression sequences, depicted the orbital optic nerve-sheath complex in its full length with a pixel resolution of 0.16 × 0.16 mm. At the same time, the image-acquisition time (11 seconds per slice) was decreased, thus reducing the risk of potential motion artifacts.
For the control, we assessed the interobserver and intraobserver reproducibility and variability to ascertain the quality of the image analyses. In our evaluation, the relatively high ICC (≥0.84) and low difference (≤0.23 mm) suggested that the standardized region-of-interest evaluation was sufficiently reliable and reproducible. [fig_ref] Figure 3: See legend on next page [/fig_ref] Oblique magnetic resonance image of the optic nerve/sheath complex (coronal T 2 -weighted fast-recovery fast spin-echo sequence (T2WI-FRFSE) with fat suppression; digital field of view = 4, window width = 2,000, window level = 1,000), taken at 3 mm [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref] , B1, C1, and D1), at 9 mm [fig_ref] Figure 2: Scheme to demonstrate the optic nerve/sheath complex [/fig_ref] , B2, C2, and D2), and at 15 mm [fig_ref] Figure 3: See legend on next page [/fig_ref] , B3, C3, and D3) behind the globe. [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref] to 3: A 70-year-old woman with spontaneous intracranial hypotension; lumbar ICP value was 3.7 mm Hg. [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref] to 3: A 39-year-old man with head injury; lumbar cerebrospinal fluid pressure was 8.5 mm Hg. [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref] to 3: A 33-year-old man with peripheral neuropathy; lumbar cerebrospinal fluid pressure was 16.2 mm Hg. [fig_ref] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve [/fig_ref] Our study confirmed previous investigations [bib_ref] Ocular sonography in patients with raised intracranial pressure: the papilloedema revisited, Geeraerts [/bib_ref] [bib_ref] Optic nerve sonography in the diagnostic evaluation of adult brain injury, Soldatos [/bib_ref] [bib_ref] Using MRI of the optic nerve sheath to detect elevated intracranial pressure, Kimberly [/bib_ref] [bib_ref] Effect of intracranial pressure on the diameter of the optic nerve sheath, Watanabe [/bib_ref] [bib_ref] Decreased diameter of the optic nerve sheath associated with CSF hypovolemia, Watanabe [/bib_ref] [bib_ref] MR imaging of the optic nerve sheath in patients with craniospinal hypotension, Rohr [/bib_ref] [bib_ref] Highresolution magnetic resonance imaging of the intraorbital optic nerve and subarachnoid space..., Mashima [/bib_ref] [bib_ref] Validation of the optic nerve sheath response to changing cerebrospinal fluid pressure:..., Hansen [/bib_ref] [bib_ref] Use of T2-weighted magnetic resonance imaging of the optic nerve sheath to..., Geeraerts [/bib_ref]. The anatomic basis for our results was the observation of free communication of CSF between the intracranial cavity and the orbital space through the optic nerve canal [bib_ref] Fast and quantitative high-resolution magnetic resonance imaging of the optic nerve at..., Weigel [/bib_ref]. The physiological explanation for our results was that the pressure in the orbital subarachnoid space is correlated with the ICP, and that the orbital subarachnoid space can distend, depending on its pressure, because of the principles of elasticity, according to the Poisson effect. Correspondingly, patients with elevated ICP had a wide orbital CSF space, whereas patients with intracranial hypotension showed a shallow orbital CSF space [bib_ref] Ocular sonography in patients with raised intracranial pressure: the papilloedema revisited, Geeraerts [/bib_ref] [bib_ref] Optic nerve sonography in the diagnostic evaluation of adult brain injury, Soldatos [/bib_ref] [bib_ref] Using MRI of the optic nerve sheath to detect elevated intracranial pressure, Kimberly [/bib_ref] [bib_ref] Effect of intracranial pressure on the diameter of the optic nerve sheath, Watanabe [/bib_ref] [bib_ref] Decreased diameter of the optic nerve sheath associated with CSF hypovolemia, Watanabe [/bib_ref] [bib_ref] MR imaging of the optic nerve sheath in patients with craniospinal hypotension, Rohr [/bib_ref]. A linear relation between invasive ICP measurements and the optic nerve-sheath diameter was reported in previous studies on patients with traumatic brain injury [bib_ref] Validation of the optic nerve sheath response to changing cerebrospinal fluid pressure:..., Hansen [/bib_ref] [bib_ref] Use of T2-weighted magnetic resonance imaging of the optic nerve sheath to..., Geeraerts [/bib_ref]. In these studies, the optic nervesheath diameter showed lower correlation coefficients (0.66 ≤ r ≤ 0.76) for the associations with lumbar CSFpressure measurements than did the OSASW in our study (0.83 ≤ r ≤ 0.88). Correspondingly, the retinal nerve fiberlayer thickness as a surrogate for the status of the optic nerve was strongly related to the optic nerve diameter (r = 0.61; P < 0.0001 at 9 mm; and r = 0.75; P < 0.0001 at 15 mm behind the globe), and the optic nerve-sheath diameter (r = 0.57, P = 0.0001 at 9 mm; r = 0.75, P < 0.0001 at 15 mm) in our study, whereas it was not related to the OSASW. It showed that the OSASW, as compared with the optic nerve-sheath diameter, was a better parameter to assess the ICP.
Our study confirmed previous investigations on the association of lumbar CSF-P measurements with body mass index and with arterial blood pressure [bib_ref] Body mass index has a linear relationship with cerebrospinal fluid pressure, Berdahl [/bib_ref] [bib_ref] Cerebrospinal fluid pressure correlated with body mass index, Ren [/bib_ref] [bib_ref] The relation between intracranial pressure, mean arterial pressure and cerebral blood flow..., Gobiet [/bib_ref]. It extends these findings to correlations between arterial blood pressure and body mass index and the OSASW.
The results of our study may have clinical implications. In patients with a presumed abnormal ICP, an MRI examination of the orbit with determination of the Stepwise multiple linear regression analysis with lumbar CSF-P measurements as the dependent variable in the training group (n = 42) A stepwise multiple linear-regression procedure was conducted to develop mathematical functions of noninvasive ICP assessment, lumbar cerebrospinal fluid pressure measurements as a dependent variable, orbital subarachnoid space width at 3 to 15 mm behind the globe, body mass index, and mean arterial blood pressure as independent variable at the same time. BMI body mass index, lumbar CSF-P lumbar cerebrospinal fluid pressure, MABP mean arterial blood pressure, OSASW03-15, orbital subarachnoid space width at 3 to 15 mm behind the globe. Multivariate analyses were constructed to assess the associations between lumbar CSF-P measurements and orbital subarachnoid space width, lumbar cerebrospinal fluid pressure, and body mass index, which were significantly associated with CSF-P in univariate analysis. BMI body mass index, Lumbar CSF-P lumbar cerebrospinal fluid pressure, MABP mean arterial blood pressure, OSASW orbital subarachnoid space width.
OSASW may be helpful to derive a hint for the ICP, provided that free communication exists between the intracranial CSF space and the orbital CSF space. From a practical point of view, the technique described in our study might be of particular interest, if a patient in a medical-emergency situation underwent a cranial MRI examination including the orbit for other reasons, so that the same image taken for information about the brain or head can simultaneously serve to find an estimation of the ICP. This may also be important for an early diagnosis of an elevated ICP, because one may assume that, in patients with acutely elevated ICP, the orbital subarachnoidal space may dilate earlier than papilledema of the optic nerve head develops.
Another clinical impact of the technique described in our study may be for patients with a chronic disease. Recent studies suggested that patients with normal-(intraocular) pressure glaucoma may have a low ICP and thus a narrow OSASW [bib_ref] Orbital cerebrospinal fluid space in glaucoma: the Beijing Intracranial and Intraocular Pressure..., Wang [/bib_ref] [bib_ref] Cerebrospinal fluid pressure in glaucoma: a prospective study, Ren [/bib_ref] [bib_ref] Intraocular pressure correlates with optic nerve sheath diameter in patients with normal..., Pinto [/bib_ref]. As a corollary, the increased ICP in patients with idiopathic intracranial hypertension may be estimated by an orbital MRI examination showing a dilated orbital CSF space. Applying the technique described in our study may thus be diagnostically helpful also for patients with a chronic disease associated with an abnormal ICP.
Although our study included 72 patients, only eight of them had an ICP >20 mm Hg. It may suggest that the results can primarily be applied only to patients without increased ICP and that the results give a hint of the ICP estimation in patients with markedly elevated ICP. Moreover, it should be noted that all CSF-P values in the present study were less than 26.5 mm Hg. Because patients with marked high ICP were not included and because the elasticity module of the orbital optic nerve meninges have not been tested, the possibility exists that the relation between CSF-P and OSASW is not linear but shows a plateau in the case of a higher ICP, particularly more than 30 mm Hg.
However, Hansen and colleagues [bib_ref] Dependence of the optic nerve sheath diameter on acutely applied subarachnoidal pressure:..., Hansen [/bib_ref] investigated the acute pressure-dependent behavior of the optic nervesheath diameter in vitro after controlled application of incremental pressure steps in the OSASW. They found that the relation of step magnitude and corresponding ONSD changes was nearly linear within a wide range of 5 to 65 mm Hg (r = 0.94; P < 0.01) [bib_ref] Dependence of the optic nerve sheath diameter on acutely applied subarachnoidal pressure:..., Hansen [/bib_ref]. Even so, when lumbar CSF-P is more than 30 mm Hg, the exact pressure-OSASW relation should be tested in future clinical study.
Potential limitations of our study should be mentioned. First, the MRI examination and the lumbar puncture were not performed at the same time. The MRI examination was carried out in the supine position 24 to 48 hours before the lumbar puncture, which was performed in the left lateral decubitus position. Because both parameters, the OSASW and the lumbar CSFpressure, show interday variations, the correlations between OSASW and lumbar CSF-pressure measurements might have been higher had both examinations been performed at the same time.
Second, besides OSASW, body mass index, and arterial blood pressure, other factors may be associated with ICP and may have to be incorporated into the formulas for the calculation of ICP.
Third, the study included Han Chinese, and other ethnic groups could differ in the weighting functions to predict the ICP. Therefore, it remains to be determined whether our study results can be applied to patients of other ethnicity.
Fourth, three-dimensional volumetric measurements of the orbital subarachnoid space compared with the two-dimensional measurements as taken in our study may have resulted in higher correlations with the lumbar CSF-P measurements.
Fifth, the method depends on the pathway from the intracranial to the orbital portion of the subarachnoid space of the optic nerve. If that pathway in the optic nerve canal or at the inner aperture of the optic nerve canal is blocked (for example, by a suprasellar meningioma, by circular adhesions as a sequel of a tuberculous meningitis, in patients with obstructive hydrocephalus, or in patients with an intracanalicular ophthalmic artery aneurysm), the orbital CSF-P (and thus the width of the orbital CSF space) is not related to the ICP [bib_ref] Sphenoid and optic nerve sheath meningioma revealed by recurrent brain infarctions, Mourad [/bib_ref] [bib_ref] Fusiform intracanalicular ophthalmic artery aneurysm; case report and review of literature, Choi [/bib_ref].
Sixth, it should be stressed that, based on the MRI calculations, the CSF-P as measured by lumbar puncture could vary by a factor of 2, so that it remains unclear whether the MRI calculations may or may not supplant lumbar-puncture measurements in the evaluation of patients with suspected abnormalities of the CSF-P. The MRI technique may be of value in choosing patients who may need invasive monitoring and the follow-up of such patients.
Future studies may address the question by which means this variation can be further reduced, and with which technique the lumbar-puncture method of the MRI measurement of the orbital CSF space may be more valid to assess the CSF-P.
# Conclusions
In patients with normal, moderately decreased, or elevated intracranial pressure, the latter could noninvasively be estimated based on MRI-assisted OSASW measurements with adjustment for body mass index and mean arterial blood pressure. If confirmed and further refined by future investigations, this finding could open the possibility to measure the intracranial pressure noninvasively. Although this technique will not replace invasive ICP monitoring, it might be helpful and a supplement in choosing patients who may need an invasive examination.
## Key message
The intracranial pressure can be estimated noninvasively, based on measurements of the width of the orbital cerebrospinal fluid space.
Abbreviations BMI: Body mass index; CI: confidence interval; CSF: cerebrospinal fluid; CSF-P: intracranial CSF pressure; FRFSE: fast-recovery fast spin-echo sequence; ICC: intraclass correlation coefficient; ICP: intracranial pressure; MABP: mean arterial blood pressure; MRI: magnetic resonance imaging; OSASW: orbital subarachnoid space width; T2WI-FRFSE: T 2 -weighted fast-recovery fast spinecho sequence.
## Competing interests
The authors declare that they have no competing interests. The intraclass correlation coefficient (ICC) was calculated to determine the degree of interobserver and intraobserver reliability. The intraobserver and interobserver variabilities were determined by Bland-Altman analysis. CI confidence interval, LoAs limits of agreement. a Based on the first measurements of the first observer and measurements of the second observer. b Based on the first and second measurements of the first observer.
[fig] Figure 1: Magnetic resonance imaging scan of the retrobulbar optic nerve ((A) transversal section, and (B) oblique sagittal section) T2WI-FRFSE of the retrobulbar optic nerve (digital field of view = 8, window width = 2,000, window level = 1,000). These images were used to plan the position of three slices at 3, 9, and 15 mm behind the globe of T2WI-FRFSE fat-suppressed sequences to measure the optic nerve diameter, optic nerve sheath diameter, and width of the orbital subarachnoid space at 3, 9, and 15 mm behind the globe. Windows, version 20.0; IBM-SPSS, Chicago, IL, USA) and the MedCalc program (version 11.5.1.0 for Windows; www.medcalc.be; accessed: 2011-4-20) [/fig]
[fig] Figure 2: Scheme to demonstrate the optic nerve/sheath complex. (A) Oblique coronal T 2 -weighted fast-recovery fast spin-echo sequence (T2WI-FRFSE) image with fat suppression for demonstrating the optic nerve/sheath complex taken at 3 mm behind the globe perpendicular to the optic nerve axis (digital field of view = 4, window width = 2,000, window level = 1,000). The nerve parenchyma is the hypodense area inside of the hypertense ring of cerebrospinal fluid. (B) Schematic drawing of the optic nerve/sheath complex including the optic nerve (the black area represents optic nerve), surrounding cerebrospinal fluid space (white rim area), and the optic nerve sheath (at the conjunction of the cerebrospinal fluid space rim and the orbital fat). OND, optic nerve diameter; ONSD, optic nerve sheath diameter. [/fig]
[fig] Figure 3: See legend on next page.) [/fig]
[fig] Figure 4: Scattergram showing the distribution of lumbar cerebrospinal fluid pressure measurements versus the width of the orbital subarachnoid space measured at 9 mm behind the globe. [/fig]
[fig] Figure 5: Bland-Altman graph of the inter-method agreement of the non-invasive magnetic resonance imaging (MRI) derived intracranial pressure (ICP) assessment using weighting function (1) (A), (2) (B), and (3) (C) and invasive lumbar cerebrospinal fluid pressure (CSF-P) measurements in the test group. CSF-P, cerebrospinal fluid pressure; ICP, intracranial pressure; SD, standard deviation. The solid line represents the mean of the difference between both methods; the two dashed lines represent the mean of the difference plus or minus the 1.96-fold standard deviation of the difference. [/fig]
[table] Table 3: Interobserver and intraobserver repeatability of the optic nerve-sheath complex measurements by MRI Optic nerve-sheath diameter [/table]
|
Social, Emotional, and Existential Loneliness: A Test of the Multidimensional Concept
Background and Objectives:Since the 1980s, most researchers have agreed on the concept of social and emotional loneliness as an unacceptable and negatively experienced discrepancy between realized and desired interpersonal relationships. For other researchers, existential loneliness stems from the realization that a human being is fundamentally alone, with the accompanying emptiness, sadness, and longing. This article examines whether instruments to measure these conceptualizations indicate a multidimensional concept.Research Design and Methods:The 2019 observation of the Longitudinal Aging Study Amsterdam (N = 1,316; aged 61-101 years; 52% women) included five direct questions about loneliness, the 11-item de Jong Gierveld social and emotional loneliness scale, and 14 items from the translated Existential Loneliness Questionnaire. Confirmatory factor analysis was conducted in Mplus. Results: Five factors were observed: direct questions, social and emotional loneliness, and loneliness in relationships and meaninglessness in life. The intercorrelations among all five factors were positive. Emotional loneliness correlated most strongly with direct questions. Discussion and Implications: Loneliness is multifaceted and means that one is not embedded in a personal network, misses closeness and intimacy, and lacks meaning in life. The emotional loneliness items most closely represent what people mean when they report loneliness.
Since the 1980s, most researchers have agreed on the concept of loneliness as a feeling of deprivation related to interpersonal relationships. defined loneliness as "the unpleasant experience that occurs when a person's network of social relations is deficient in some important way, either quantitatively or qualitatively." Other researchers have defined existential loneliness "as an intolerable emptiness, sadness, and longing, that results from the awareness of one's fundamental separateness as a human being" . The two definitions and, as will be shown, their measuring instruments not only use similar terms, but also highlight a lack of social connection, albeit in a very different way. This article examines whether the instruments for measuring these conceptualizations indicate a multidimensional concept.
## The concepts of social, emotional, and existential loneliness
Loneliness is a negative feeling related to loss and disappointment. It is the outcome of a process in which a person weighs up his or her existing personal relationships against his or her own wishes and social expectations with regard to relationships. If the social network is too small or the relationships are of insufficient quality, the person often feels lonely. This has been the most widely used conceptualization of loneliness in research. Two basic types of loneliness are social and emotional loneliness. Social loneliness originates from the absence of a broader group of contacts or an engaging social network. Emotional loneliness originates from the absence of an intimate figure or a close emotional attachment. Social and emotional loneliness often occur due to reduced social activitiesor loss of the partner, among other situations.
In a literature review,distinguished several key aspects of existential loneliness: not connecting with others and the world outside, alienation, feelings of isolation, emptiness, and abandonment. Additionally, mortality-related fears were identified to be associated with this type of loneliness, including the fear of disappearing from the earth, the fear of being forgotten, and the fear of dying. A further examination of this concept leads to two considerations.
First, the concept of existential loneliness needs to be sharpened and more clearly distinguished from social and emotional loneliness. For example, a negative evaluation of not connecting with others overlaps with social loneliness. Feelings of isolation, emptiness, and abandonment are aspects of emotional loneliness.
Second, social and emotional loneliness happen to somebody; people do not seek loneliness voluntarily. These types of loneliness are felt as social pain. In contrast, some scholars see existential loneliness as something that can be appreciated positively; one finds strength in being solitary, although it takes effort to realize this situation. For, existential loneliness is inherent in human existence, such that everyone can identify as a solitary individual. People retreat from the social world to create a self-identity and to live a genuine life. Thus, in life's most intimate and important moments, someone is an individual and is always alone. Loneliness can be a creative force among people who create music or inventions on their own.has described the separation by which humans give meaning to life and to their relationship with God. For, existential loneliness consists of negative feelings of nothingness that alternate and amalgamate with being in a process of inner growth. In this article, we focus on existential loneliness as a negative experience, and we aim to use the concept of solitude for when people want to be alone with positive purposes.
Existential loneliness is often studied in patients with severe illnessand end-of-life situations.studied older people residing in nursing homes. After living a fulfilling and often busy life, these people are left without meaningful roles.studied people with severe illnesses who realize their own mortality.examined loneliness in AIDS patients and observed that patients experienced inescapable separateness from others. Existential loneliness differs from social and emotional loneliness in two ways. First, social and emotional loneliness are associated with a lack of meaningful social relationships and a lack of social companionship. Existential loneliness is the result of a broader separation related to the nature of existence and, in particular, a lack of meaning in life. An individual may be in the company of others but experience existential loneliness. Second, social and emotional loneliness can be overcome by improving the quality of the network of relationships or by adjusting the level of aspiration. Existential loneliness, on the other hand, has no permanent remedy according to the phenomenological approach.
## The measurement of loneliness
Loneliness is often measured with a single, direct question. Such a question, for example, "Do you feel lonely?," is simple to use, appears to be acceptable to respondents, reflects loneliness as understood by the respondent, and provides an easy way to assess the prevalence of loneliness. However, the use of a direct question presupposes that the respondents have a common understanding of the term "loneliness" and that their understanding encompasses the whole theoretical concept. A single item does not provide information on the relevance of social, emotional, and existential aspects of loneliness. Because of the social stigma of loneliness, people who are not seen as lonely by others may find it difficult to admit their loneliness as an answer to a direct question. Finally, the psychometric quality of a single question cannot be determined.
Alternatively, loneliness can be measured with a scale including statements that relate to loneliness but avoid the term "loneliness" or similar wording. The UCLA scaleand the de Jong Gierveld (DJG) scaleare based on a conceptual framework of loneliness, in which different relational aspects and emotions relevant to the experience of loneliness are distinguished.concluded that the UCLA scale appeared to identify a subjective lack of social companionship and is less sensitive to philosophically determined types of loneliness, such as existential loneliness.andfound various dimensions in the UCLA scale, but it is often considered a unidimensional measure. The DJG scale was developed as a unidimensional loneliness scale with both the social and emotional aspects of loneliness in mind. The homogeneity of the unidimensional scale proved to be modest at best. When searching for more homogeneous subscales, social and emotional loneliness factors emerged. The UCLA and DJG scales have rarely been studied in one sample. In a Dutch study,observed that the measurement of social loneliness was similar to that of the core dimension of the UCLA scale, which includes seven items. Another studyindicated that both scales were multidimensional, but the correlation between the scale scores was not reported.
While scales for social and emotional loneliness are widely used, there has been limited research into existential loneliness.developed the Existential Loneliness Questionnaire (ELQ). This is the only scale we found that focuses on existential loneliness in survey research, with the exception of one studied among students in the 1970s. The ELQ consists of 22 items and was found to be sufficiently internally consistent. Three items were specific to HIV, three were formulated conditionally, and two included a direct reference to loneliness. Gökdemir-Bulut and Bozo (2018) conducted two studies on the ELQ. They found three factors, that is, loneliness in social ties, loneliness in close relationships, and finding meaning in life; it appears that the first and second factors relate to social and emotional loneliness.
## The current study
As far as we know, no research has examined whether existential loneliness should be distinguished from social and emotional loneliness. The current research analyzes data from older respondents who answered items from the DJG and ELQ scales, along with direct loneliness questions. It tests whether social, emotional, and existential loneliness are related but distinctive dimensions of loneliness that are associated with direct measurements of loneliness. We also aim to determine whether existential loneliness is indeed a negative feeling.
To support the validity of the loneliness scales, we look for the association of various characteristics with the three types of loneliness. We expect to find that men and women have similar levels of loneliness scales scoresbut that women will report more loneliness on the direct questions. We also expect that age will correlate positively with all instruments, as will being without a partner, having a smaller network, not having daily network contact, and having poor health because unhealthy people have little social participation. Religiosity is linked to lower levels of loneliness. Religious people are more involved than others in churchbased social contacts and organizations in the community. In the current study, validity testing can go both ways. When there is a strong core in the concept of loneliness, we can establish that measurements of social, emotional, and existential loneliness are similarly associated with a number of antecedents. When multiple dimensions in loneliness need to be distinguished, we can establish that the measurements are differently associated with the antecedents.
# Design and methods
## Respondents
Data were taken from the Longitudinal Aging Study Amsterdam. This program employed samples of men and women born between 1908 and 1957 taken from the population registers of the cities of Amsterdam, Zwolle, and Oss, and six surrounding small municipalities in 1992, with additions in 2002 and 2012. The response rate at baseline was 63%. Follow-up interviews were conducted every 3 or 4 years. For each follow-up, an average of 82% of respondents were reinterviewed.
Data collection on 1,701 men and women was completed in 2019. We excluded 110 participants for whom data were collected from an interview with a proxy and 198 respondents interviewed by phone with a short questionnaire including only six social and emotional loneliness items. The DJG scale and three direct questions were part of the first, general face-to-face interview. One respondent was unable to participate in a full interview due to incapacity and did not answer the loneliness questions. The ELQ was part of a subsequent medical interview, which was conducted by another interviewer an average of 32 days after the initial interview. A total of 76 respondents from the first interview did not participate in the second interview. Thus, data from 1,316 respondents were analyzed. Their age varied between 61 and 101 years (M = 73.0), and 52% were women. Migrants were underrepresented. Most respondents were of Dutch origin; 3% were born in another Western country, and 2% were born in a non-Western country. Almost all respondents lived independently; 1% lived in a nursing home.displays all loneliness questions and response options. Three direct loneliness questions were asked: a selfrating; an item assessing whether respondents "sometimes feel lonely"; and "During the past week I felt lonely," which is adapted from the Center for Epidemiologic Studies Depression scale. Two ELQ items that asked respondents whether they feel "lonely" and "alone" were also treated as direct questions.
## Measures
From the DJG scale, five positively phrased items and six negatively phrased items measure the intensity of deprivation, which is considered to be the essence of loneliness. The subscales reflect social loneliness and emotional loneliness, respectively. Existential loneliness was measured by 14 ELQ items. We excluded three HIV items and three conditionally formulated items. We performed a translation from English into Dutch, followed by back translation, using two translators in both steps. Marital and partner status was assessed by three questions. Network members with whom there was regular contact and who were important to the respondent were identified by name across seven domains. We derived two characteristics: personal network size (not counting the partner) and being in daily contact with someone in the network. Self-rated health was measured using the question "How is your health in general?" (poor to very good). The frequency of church attendance was assessed on a six-point scale from "never" to "once a week or more."
## Procedure
The scores on the five direct questions were recoded so that the answers "not lonely," "rarely or never," "no," and "no!" indicate the absence of loneliness and the other answers indicate loneliness; it was assumed that the middle category "more or less" indicates loneliness. We investigated the psychometric properties of the scales. Homogeneity (H), computed by Mokken scale analysis, was indicated by the correlation between the item scores (scale lower limit is 0.30, indicating weak homogeneity; 0.40-0.50 indicates medium homogeneity; 0.50 or higher shows strong homogeneity; Sijtsma & van der Ark, 2017).
Reliability describes the interrelationship in terms of the number of items (lower limit 0.80;.
We applied confirmatory factor analysis in Mplus. Item scores were treated as having categorical (direct questions) and ordinal (all other) measurement levels; latent variables were continuous. First, we tested whether there are five factors underlying the 30 items. The five factors are direct measurements, social loneliness, emotional loneliness, and the existential aspects "loneliness in relationships" and "meaninglessness in life," which are derived from the work of Gökdemir-Bulut and Bozo (2018). Next, to investigate the commonality of the five factors, we tested a unidimensional, second-order, and bifactor model. To test the model fit, we used χ 2 /df, the comparative fit index (CFI), the Tucker-Lewis index (TLI), and the root mean square error of approximation (RMSEA): χ 2 /df <3, CFI and TLI ≥0.95, and RMSEA ≤0.07 indicate an acceptable model fit. The model was estimated using the full information maximum likelihood method. We present unstandardized factor loadings herein. Correlations between factors were corrected for attenuation. Estimated factor scores were derived to create individual loneliness scale scores. We also computed Pearson correlations between loneliness scale scores and the antecedents of loneliness. Differences in strength between loneliness scores were tested with the z-statisticAsked in the first interview. Response options: "not lonely," "moderately lonely," "strongly lonely," and "very strongly lonely." h Asked in the first interview. Response options: "no," "more or less," and "yes." i Asked in the first interview. Response options: "rarely or never," "some of the time," "occasionally," and "mostly or always." j Asked in the second interview. Response options: "no!," "no," "more or less," "yes," and "yes!"
# Results
There were few missing data. One respondent did not answer five items, and 16 respondents did not answer ELQ item 16. The frequency of loneliness as indicated by the 30 items varied considerably. Of the three direct questions in the first interview, between 19% and 25% of the respondents reported being lonely. Only approximately 10% reported loneliness using the two direct ELQ questions. For the social loneliness items, an average of 17% of respondents answered affirmatively to indicate loneliness (including answers in the middle category). Many respondents answered negatively to the item "There are many people I can trust completely" (DJG 7). On average 15% reported emotional loneliness. Few respondents answered affirmatively to the item "I often feel rejected." On average 20% reported existential loneliness in relationships. Compared to other items, respondents answered more affirmatively on ELQ item 12, that is, "can trust and rely," indicating low levels of loneliness; however, a differently worded item on trust from the DJG scale (item 7) indicated higher levels of loneliness. There were a high number of affirmative responses to the item "Important relationships have ended or become weaker." Finally, the average for meaninglessness in life was 19%. There were few affirmative answers to the item "I feel dead." Many respondents disagreed with the statement "There is a purpose in my life." The homogeneity of loneliness as measured by direct questions was strong (0.75;, and the reliability was sufficient (Kuder-Richardson 20 [KR-20] = 0.86). For social and emotional loneliness, the homogeneity was strong (0.51 and 0.53, respectively), and the reliability was (almost) sufficient (KR-20 = 0.79 and 0.84, respectively). The two existential loneliness scales are weakly homogeneous (0.30 and 0.34) and have low reliability (Cronbach's alpha = 0.69 and 0.73).
The five-factor model had the best fit and three of the four indices indicated that the model fit the data. For each factor, one item served as a reference with a factor loading of 1; for the direct questions factor, the reference item was the self-rated item. The lower estimated factor loadings for three items ("sometimes lonely," "last week lonely," and "feel alone") indicate a weaker correlation between the item and the latent factor compared to the other two items. The R 2 represents these correlations.
The high item-factor correlations for direct questions and social and emotional loneliness factors and the lower correlations for the two existential loneliness factors are congruent with the differences in homogeneity and reliability reported previously.
Attenuation-corrected correlations between the five loneliness scale scores varied between 0.32 and 0.82. The highest correlation was found between the direct questions and emotional loneliness. Correlations among the two existential loneliness scales and the other factors were relatively low.
Most respondents (63%) lived with their spouse; fewer than 1% lived independently while their spouse lived in a nursing home; 3% were not married and lived with a partner; 5% had a partner who lived outside of the household; and 29% did not have a partner. Most respondents had extensive personal networks: fewer than 1% had no or only one tie, and 91% had six or more ties; the average network size was 16.5 ties. In addition to daily contact with the partner with whom one lived (66%), 18% had daily contact with someone else, and 15% had no daily network contact. Health was rated as good or very good by 68%. The frequency of church attendance was monthly or more frequently for 23% of the respondents. Correlations between the loneliness scales and gender (with the exception of emotional loneliness), age, network characteristics, and health were in the direction similar to the directions reported in the literature. The correlations were generally modest and there were few differences between the five scales. Most striking was that the correlation with living with spouse or partner was relatively high for the direct questions and emotional loneliness. Loneliness did not correlate with the frequency of church attendance.
# Discussion and implications
By applying the translated ELQ among older Dutch adults, we found evidence for multiple dimensions of loneliness in addition to social and emotional loneliness. Gökdemir-Bulut and Bozo (2018) tested the ELQ and assumed three factors. Because we excluded a number of items for different reasons, we combined two factors into one dimension, namely, "existential loneliness in relationships." This ELQ dimension focuses on social bonding in relationships and has similarities with social loneliness (e.g., item 12 with DJG item 7). However, this ELQ dimension was only modestly related to social loneliness. This modest association may be due to addressing different aspects of loneliness in the other items and due to different wordings and mode effects, such as the use of two different interviews and different numbers of response options. The congruence of "existential loneliness in relationships" and social and emotional loneliness needs to be further investigated. We conclude that the ELQ subscale on relationships does not contribute sufficiently to the conceptualization of loneliness. In addition, the subscale is weakly homogeneous and insufficiently reliable. The third factor distinguished by Gökdemir-Bulut and Bozo relates to a differential conceptualization that we have labeled "meaninglessness in life." The positive correlations among existential loneliness scale scores, the direct question scores, and social and emotional loneliness show that existential loneliness as measured with the ELQ is not a positive and rewarding experience. However, we did not address the fact that individuals sometimes seek solitudeor think they are better off alone, and we did not measure the "desire for aloneness". The lack of fit of the unidimensional, second-order, and bifactor models and the modest correlations between the five factors, especially between social and emotional loneliness on the one hand and existential loneliness on the other, suggest that the various instruments address many different aspects. We can see the collection of these aspects as representative for one broad multidimensional theoretical concept, or, alternatively, these aspects do not have enough in common to be seen as dimensions of one sharply defined concept of loneliness and represent two concepts, namely, "loneliness" and "meaning in life", each with their own definition. The three dimensions can also be interpreted to cover three aspects of the life experience: how satisfied people are with their lives (evaluative), feelings and moods (hedonistic), and assessing giving meaning to life (eudemonic). However, unlike Steptoe et al.'s findings, we did not find clearly distinct antecedents associated with the loneliness dimensions. This may indicate a commonality: social, emotional, and existential aspects are all negative loneliness feelings pointing to a lack of connectedness with people and life, which are associated in the same way to a number of loneliness antecedents. On the basis of the antecedents, the three dimensions cannot therefore be sufficiently distinguished from each other to be treated as a standalone phenomenon. In future research, we propose using 18 questions that measure three types of loneliness. The existential scale could be improved.
We also included direct questions on loneliness in our analyses. The strong homogeneity of the factor with direct questions may be related to the repetition of the word "loneliness" (or variants) in four of the five items. Balancing the advantages and disadvantages of direct questions compared to scales, many researchers prefer the use of scales. The use of a direct question can be of additional value when a strong correlation is observed with a loneliness scale score to confirm the validity of the scale. The high correlation we observed between the scores for the direct questions and the emotional loneliness scale scores suggests that the evaluations and feelings included in the six emotional loneliness items most closely represent what people mean when they report loneliness. A researcher who prefers to use only one direct question does not know which type of loneliness the question measures, or in the best case, it most likely measures emotional loneliness in an unreliable way.
It is now widely recognized that loneliness is related to public health, for example, due to increased premature mortality among lonely people. Therefore, policy makers are interested in the prevalence of loneliness, and practitioners treating clients who may be lonely need to assess whether loneliness is the problem. Scale ratings of loneliness indicate the severity of loneliness but do not indicate the prevalence of loneliness or whether a person is lonely. Providing a cutoff score of a loneliness scale may solve this problem but also needs a reference that can be found in answers on direct loneliness questions. We found that the frequency of signs of loneliness varies greatly between items. This may have to do with the period covered in the items (e.g., last week or undefined), the wording of the items (e.g., trusting many people completely or having a trustee), the different number of response options (three to five), and the context of the interview (i.e., general or medical). This suggests that both the basis for determining loneliness prevalence and the likelihood of an individual being assessed as lonely vary based on the design of the research. However, we observed strong homogeneity among the answers to the five direct questions. This offers opportunities for an improved way to estimate the prevalence of loneliness and to assess individuals. Answers to various questions can be averaged, and this average can be used to determine appropriate cutoff scores.
We want to draw attention to some design issues in addition to those addressed previously. First, respondents participated in two long interviews. The sample does not include many older people who are in a serious situation, such as severe illness or the end of life, which is the focus of a number of studies on existential loneliness. Second, there was a time between the two interviews, which contributed to participants avoiding response sets and other ways in which offering many questions on one topic can affect the outcome of the research. At the same time, the respondent's situation may have changed between the two interviews, which may have contributed to the relatively low correlation between the DJG scores (assessed in the first interview) and the ELQ scores (assessed in the second interview). However, we also note that the factor of direct questions, of which three were asked in the first interview and two were asked in the second interview, had very good psychometric properties. This suggests that the interval between the two interviews did not strongly influence our results. Finally, we did not look in detail at the quality of the scales and did not investigate whether removing items leads to more homogeneous scales.
To conclude, we explored the dimensions of loneliness and found that social, emotional, and existential aspects were relevant. These were equally related to the antecedents of loneliness. Direct questions may be of additional value in assessing loneliness among older adults.
# Funding
The Longitudinal Aging Study Amsterdam is supported by a grant from the Netherlands Ministry of Health, Welfare and Sport, Directorate of Long-Term Care. |
Metal Complexes of Diisopropylthiourea: Synthesis, Characterization and Antibacterial Studies
Co(II), Cu(II), Zn(II) and Fe(III) complexes of diisopropylthiourea have been synthesized and characterized by elemental analyses, molar conductivity, magnetic susceptibility, FTIR and electronic spectroscopy. The compounds are non-electrolytes in solution and spectroscopic data of the complexes are consistent with 4-coordinate geometry for the metal(II) complexes and six coordinate octahedral for Fe(III) complex. The complexes were screened for their antibacterial activities against six bacteria:Escherichia coli, Pseudomonas auriginosa, Klebsiella pneumoniae, Bacillus cereus, Staphylococcus aureus and Bacillus pumilus. The complexes showed varied antibacterial activities and their minimum inhibitory concentrations (MICs) were determined.
# Introduction
Substituted thioureas are known to form stable, neutral coordination compounds with a variety of transition metal ions and some has been structurally characterized. The chemistry of substituted thiourea derivatives has attracted attention because of their potential use as reagents for the separation of metal ions [bib_ref] Intramolecular hydrogen-bond controlled unidentate coordination of potentially chelating N-acyl-N'-alkylthioureas: crystal structure of..., Bourne [/bib_ref] and as antibacterial [bib_ref] Some 3-thioxo/Alkylthio-1,2,4-triazoles with a substituted thiourea moiety as possible antimycobacterials, Kuzukguzel [/bib_ref] , antiviral [bib_ref] Regiospecific synthesis, X-ray crystal structure and biological activities of 5-bromothiophenethyl thiourea, Venkatachalam [/bib_ref] and antifungal agents [bib_ref] Thiourea derivatives and their nickel(II) and platinum(II) complexes: Antifungal activity, Campo [/bib_ref]. Apart from their various applications, these ligands are of interest because they have three potential coordination sites: the sulfur of the CS group and the nitrogen atom of the NH groups. Infectious diseases are a major cause of death OPEN ACCESS especially in developing countries [bib_ref] Antiplasmodial and antitrypanosomal activities of aminocyclo[2.2.2]octyl ω-aminoalkanoates, Schlapper [/bib_ref] [bib_ref] The economic burden of malaria, Gallup [/bib_ref]. The development of antimicrobial drugs, particularly antibiotics, has long being touted as one of the great medical success stories of the twentieth century [bib_ref] Malaria and the red blood cell membrane, Cook [/bib_ref]. At present, resistance against antimicrobial agents is becoming a public health problem worldwide [bib_ref] New directions in antibacterial research, Chu [/bib_ref] [bib_ref] Surveillance uncovers the smoking gun for resistance emergence, Verhoef [/bib_ref]. In the search for novel therapy, recent research [bib_ref] In vitro and in vivo antimalarial activity of ferrocholoquine, a ferrocenyl of..., Delhaes [/bib_ref] [bib_ref] Synthesis, characterisation and antiplasmodial studies of metal(II) complexes of sulfadiazine, Ajibade [/bib_ref] [bib_ref] Toward a novel metal-based chemotherapy against tropical diseases.7. Synthesis and in vitro..., Navarro [/bib_ref] [bib_ref] Transition metal complexes of buparvaquone as potent new antimalarial agents: Synthesis, X-ray..., Gokhale [/bib_ref] has demonstrated that attaching organic drugs to metal containing fragments can enhance their activity. As part of our studies [bib_ref] Synthesis, characterization and antiprotozoal studies of some metal complexes of antimalarial drugs, Ajibade [/bib_ref] [bib_ref] Cobalt(III) complexes of antimalaria drugs: Synthesis, characterization and in vitro antiprotozoal studies, Ajibade [/bib_ref] [bib_ref] Antibacterial activity of metal complexes of antifolate drug pyrimethamine, Idemudia [/bib_ref] [bib_ref] Synthesis, characterisation and antiplasmodial studies of metal(II) complexes of sulfadiazine, Ajibade [/bib_ref] [bib_ref] Synthesis, characterization, antiplasmodial and antitrypanosomal activity of some metal(III) complexes of sulfadiazine, Ajibade [/bib_ref] [bib_ref] Synthesis, characterization and in vitro antiprotozoal studies of iron(III) complexes of some..., Ajibade [/bib_ref] [bib_ref] Synthesis, characterization and antibacterial activity of the metal complexes of phenylthiourea: the..., Ajibade [/bib_ref] to develop metal based therapeutic agents, we report the synthesis, characterization and antimicrobial screening of metal complexes of diisopropylthiourea.
# Results and discussion
Cobalt(II), Copper(II), Zinc(II) and iron(III) complexes of diisopropylthiourea has been synthesized by the reaction between the metal salts and diisopropylthiourea in ethanol. The complexes are air stable, non-electrolytes in solution and were characterized by elemental analysis, UV-Vis, FTIR, magnetic susceptibility measurements, conductivity, and M.pt/decomposition temperature. The zinc(II) complex, [ZnCl 2 (diptu) 2 ] was further characterized by single crystal X-ray crystallography. The analytical and spectroscopic data for the complexes are presented in [fig_ref] Table 1: Analytical data for metal complexes of diisopropylthiourea [/fig_ref] and their proposed structures in . The Zn(II), Co(II) and Cu(II) complexes formed four coordinate species by coordinating to two molecules of diisopropylthiourea through the sulfur atom and two chloride ions while the Fe(III) complex coordinated with three molecules of diisopropylthiourea and the octahedral geometry around the metal ion was completed by coordinating to three chloride ions. Single crystals suitable for X-ray analysis was obtained from methanol/dichloromethane mixtures by slow evaporation at room temperature. The crystals were air stable and moisture stable. The molecular structure of the complex along with atom numbering scheme is given in [fig_ref] Figure 3: Molecular structure of [ZnCl 2 [/fig_ref] , parameters for crystal data collection and structure refinements are in [fig_ref] Table 2: Summary of crystal data and structure refinement for [ZnCl 2 [/fig_ref]. Selected bond lengths and angles are presented in and the hydrogen bonding matrix is shown in . The complex, [ZnCl 2 (diptu) 2 ] crystallizes in orthorhombic, space group Pbcn and the geometry around the zinc atom is best described as a distorted tetrahedral [ZnS 2 Cl 2 ] environment. The tetrahedral coordination around the zinc atom consists of two monodentate diisopropylthiourea ligands and two chloride ions at the apices [fig_ref] Figure 3: Molecular structure of [ZnCl 2 [/fig_ref]. The orientation of the diisopropylthiourea ligands, defined by M = Co, Cu, Zn Cl-Zn-S-C and Zn-S-C-N torsion angles, is determined by intramolecular N-H···Cl hydrogen bonding interaction. (16) Å] for terminally coordinated thiourea [bib_ref] Supplement Tables of bond lengths determined by X-ray and neutron diffraction. Part..., Orpen [/bib_ref]. In the Zn(II) complex, the C-N and C-S bonds shows intermediate bond length between single and double bond due to the delocalization of electron in the amide bond. The two Zn-Cl bond distances are similar to each other, as are the Zn-S bond distances. The Zn-S bond show no significant differences from the Zn-S bond distances in other known zinc-sulfur complexes in a pseudo-tetrahedral environment [bib_ref] N-H···O hydrogen bonding interactions in tetrahedral [ZnS4] complexes of relevance to zinc..., Docrat [/bib_ref].
## Bond length bond angles
[formula] Zn(1)-Cl(1) 2.2634(4) Cl(1)-Zn(1)-Cl(1)#1 112.85(2) Zn(1)-Cl(1) [/formula]
The packing of [ZnCl 2 (diptu) 2 ) 2 ] molecules in the crystal lattice [fig_ref] Table 4: Distances [/fig_ref] consists of parallel chains formed by intermolecular H-bonding. Each discrete molecule interacts likewise with four neighboring molecules. For the intermolecular H-bond, each molecule of the ligand acts as an H-bond donor as well as an acceptor. The two dimensional hydrogen bonding network observed in the structure are intermolecular hydrogen bond, viz. N(1)-H(1)…Cl(1)#2, and intramolecular hydrogen bond, viz. N(2)-H(2)…Cl(1). As a result of the intermolecular hydrogen bond, one molecule of the complex is linked to four adjacent molecules [fig_ref] Table 4: Distances [/fig_ref]. The NH groups of the diisopropylthiourea ligands are arranged such that they facilitate the formulation of intra-and inter-molecular hydrogen bonds, involving Cl-anions as acceptors in both cases . This is similar to what is seen in other zinc(II) bis(thiourea) polymers [bib_ref] Dichlorobis(1,3-dimethylthiourea kappa S) zinc(II), Burrows [/bib_ref] [bib_ref] The influence of hydrogen bonding on the structure of zinc co-ordination polymers, Burrows [/bib_ref]. The intermolecular hydrogen bonds link the molecules into infinite hydrogen bonded networks
## Infrared spectra of metal complexes of diisopropylthiourea
The solid state IR spectra of diisopropylthiourea and the metal complexes in the region 4000-400 cm −1 were compared and assigned on careful comparison. In the region 3450-3100 cm −1 , the N-H vibrations that appear as a broad peak at 3450 cm −1 in the spectrum of the ligand shifted to higher wave numbers in the complexes. These shifts might be attributed to the S→M bond and increase in the C-N bonds [bib_ref] Synthesis, characterization and antibacterial activity of the metal complexes of phenylthiourea: the..., Ajibade [/bib_ref] [bib_ref] Vibrational and electronic studies on some metal thiourea complexes, El-Bahy [/bib_ref]. The band at about 1480 cm −1 in the ligand shifted to about 1486-1524 cm −1 in the spectra of the complexes. These shifts can be ascribed to increase in the double bond character of the C-N bond on complex formation. The infrared spectra of the complexes confirm that the thiourea ligands are coordinated to the metal ions via the exocyclic sulfur atom with a reduction in the π-electron density of the exocyclic C=S bond. In addition, the υ(C=S) bond of the free ligand is red-shifted to lower frequencies in the complexes.
## Electronic spectra of metal complexes of diisopropylthiourea
Co(II) complexes is the most important d 7 species known in all the coordination numbers and because of its stereochemical diversity, its spectra have been widely studied [bib_ref] Synthesis and structural studies of newMannich base ligands and their metal complexes, Al-Jeboori [/bib_ref]. In a cubic field, three spin allowed transitions are anticipated because of the splitting of the free-ion, ground 4 F term, and the accompanying 4 P term. For d 7 ions in tetrahedral crystal fields, the splitting of the free-ion, ground F term is the reverse of that in octahedral field, so that 4 A 2g (F) lies lowest. Thus spectra of cobalt (II) complexes usually consist of a broad, intense band in the visible region with a weaker one in the infrared. The 4 A 2 (F)→ 4 T 1 (F) and 4 A 2 → 4 T 1 (P) transition appear as multiple absorption in the near and visible regions, respectively. There are also bands occurring as shoulder in the high frequency side assigned to [bib_ref] Regiospecific synthesis, X-ray crystal structure and biological activities of 5-bromothiophenethyl thiourea, Venkatachalam [/bib_ref]
## A 2 (f)→ 4 t 1g (f) [ 4 a 2 (f)→ 4 t 2 (f)]. examination of this part of the infrared has sometimes indicated the presence of a band, though overlying vibrating bands makes interpretation difficult.
The electronic spectrum of [CoCl 2 (diptu) 2 ] showed two bands at 600 and 650 nm typical of Co(II) tetrahedral complexes. A room temperature magnetic moment of 4.0 B.M confirms the proposed tetrahedral geometry for the complex.
[CuCl 2 (diptu) 2 ] consists of bands in range 540-650 nm and 360-430 nm and were assigned to 2 B 1g → 2 A 1g and 2 B 1g → 2 B 2g (p) respectively indicating that the complex is in a square planar environment [bib_ref] Spectroscopic characterization of some tetradentate Schiff bases and their complexes with nickel,..., Mokhles [/bib_ref] [bib_ref] Synthesis and Characterization of New Schiff Bases Derived from N-(1)-Substituted Isatin with..., Ahlam [/bib_ref]. The complex has a room temperature magnetic moment of 2.10 B.M indicating a mononuclear Cu(II) complex. Generally, mononuclear Cu(II) complexes have magnetic moments of about 1.9-2.2 B.M due to orbital contribution to the magnetic moment and spin orbital contributions. The electronic spectrum of [Fe(diptu) 2 ] exhibited charge transfer transitions from 600 to 340 nm. Effective magnetic moment of 5.1 B. M. confirms that the complex is high-spin.
## Antibacterial studies of the complexes
The complexes were screened against six bacteria microorganisms: Escherichia coli (ATCC 8739), [bib_ref] Some 3-thioxo/Alkylthio-1,2,4-triazoles with a substituted thiourea moiety as possible antimycobacterials, Kuzukguzel [/bib_ref] ] complexes respectively and 11.5 mm against [CuCl 2 (diptu) 2 ]. The smallest zones of inhibition were exhibited by P. aureginosa (9 mm) for [CoCl 2 (diptu) 2 ] and 9.5 mm for both [CuCl 2 (diptu) 2 ] and [FeCl 3 (diptu) 3 ] complexes. The MIC values for the microorganisms vary between 2.5 to 5 mg/mL with the entire microorganism having MIC value of 2.5 mg/mL against [CoCl 2 (diptu) 2 ] and 5 mg/mL against [CuCl 2 (diptu) 2 ] except for E. Coli against [CuCl 2 (diptu) 2 ] with MIC value of 2.5 mg/mL. It can thus be concluded that the metal complexes have a broad spectrum antibacterial activity against all the tested microorganisms. Although the antibacterial activity can best be described as mild to moderate, the complexes can be optimized further through derivatization to enhance their antibacterial potential. . Zones of inhibition of the complexes at 10 mg/mL.
## Complex/bacteria [cucl 2 (diptu) 2 ] [cocl 2 (diptu) 2 ] [fecl 3 (diptu) 3 ] e. coli
12.5 11.5 13.0 P. auruginosa 9.5 9.0 9.5 K. pnemoniae
## Experimental section
# Materials and instrumentation
All reagents, metal salts, trimethoprim, pyrimethamine and amodiaquine were used as obtained from Aldrich. Elemental analyses were performed at the Micro-analytical laboratory of the School of Chemistry, The University of Manchester, UK. IR spectra were obtained as KBr discs on a Perkin-Elmer Paragon 1000 FTIR spectrophotometer equipped with CsI window (4000-250 cm −1 ). UV-Vis spectra were obtained on a Perkin-Elmer Lambda 20 spectrophotometer. Magnetic susceptibility measurements were carried out at room temperature using Sherwood scientific magnetic susceptibility balance. Diamagnetic corrections were made using Pascal's constant.
## Synthesis of metal complexes of diisopropylthioruea
## Synthesis of [fecl 3 (diptu) 2 ]
FeCl 3 ·6H 2 O (1.6218 g, 6 mmol) dissolved in 40 mL ethanol was added to a solution of 1,3-diisopropylthiourea (2.8850 g, 18 mmol) in 60 mL ethanol and refluxed for 5 h. The solution was concentrated in vacuo, filtered and the precipitate washed with ethanol, diethylether and air dried. Yield: 2.6 g (67 %).
## X-ray crystallography
Single-crystal X-ray diffraction data were collected on a Bruker KAPPA APEX II DUO diffractometer using graphite-monochromated Mo-Kα radiation (χ = 0.71073 Å). The crystal was coated with paratone-N oil and mounted on a cryoloop. Data collection was carried out at 173(2) K to minimize solvent loss, possible structural disorder and thermal motion effects. Temperature was controlled by an Oxford Cryostream cooling system (Oxford Cryostat). Cell refinement and data reduction were performed using the program SAINT. The data were scaled and empirical absorption corrections were performed using SADABS. The structure was solved by direct methods using SHELXS-97and refined by full-matrix least-squares methods based on F2 using SHELXL-97 and using the graphics interface program X-Seed [bib_ref] A software tool for supramolecular crystallography, Barbour [/bib_ref]. The programs X-Seed and POV-Ray [bib_ref] Molecular graphics: From science to art, Atwood [/bib_ref] were both used to prepare molecular graphic images. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms, except H1 and H2, were positioned geometrically and refined as riding on their parent atoms, with U iso (H) = 1.2 U eq (C) and 1.5 U eq (methyl C). The positions of H1 and H2 were placed in difference Fourier maps and refined independently with simple bond length constraints. The structure was refined successfully with R factor 0.0223.
## Antibacterial studies of the metal complexes
The antibacterial activity of the metal complexes was evaluated against six bacteria obtained from the American Type Culture Collection (ATCC). The antibacterial activity of the complexes was qualitatively determined by a modified disc diffusion method. Antibacterial activity was indicated by the presence of clear inhibition zones around the discs. Substances showing positive antibacterial activity via the disc diffusion assay were subjected to the broth diffusion method for quantitative measurement of microbiostatic (inhibitory) activity [bib_ref] Discovery and Development in the New Drugs for Systematic Opportunistic Infections, Hufford [/bib_ref]. The lowest concentration that completely inhibited visible microbial growth was recorded as the minimum inhibitory concentration (MIC). Amphicillin was used as positive control for bacterial growth.
# Conclusions
Diisopropylthiourea ligand acts as unidentate ligand through the sulfur atom forming strong complexes with Co(II), Cu(II), Zn(II) and Fe(III). The complexes were characterized by elemental analyses, molar conductivity, magnetic susceptibility, FT-IR and spectroscopic techniques. The single crystal X-ray structure of [ZnCl 2 (diptu) 2 ] is also reported. Evidence from the coordination chemistry of the crystal structure of the Zn(II) complex is dominated by strong intramolecular hydrogen bonds which lock the diisopropylthiourea N-H moiety and the chloride ion. There are also intermolecular NH-Cl interactions. The antibacterial studies of three of the complexes were studied and their minimum inhibition concentrations determined.
[fig] Figure 1, Figure 2: Proposed Proposed structure for iron(III) complex of disopropylthiourea. 1. The Molecular Structure of [ZnCl 2 (diptu) 2 ] [/fig]
[fig] Figure 3: Molecular structure of [ZnCl 2 (diptu) 2 ]. The atoms of the asymmetric unit are labeled, the other half of the molecule is generated via symmetry transformation −x, y, −z + 1/2. [/fig]
[fig] Table 4: Distances (Å) and angles (°) involving hydrogen bonding of [ZnCl 2 (diptu) 2 ]. used to generate equivalent atoms: #1: −x, y, −z + 1/2; #2: −x + 1/2, y + 1/2, z. The Cl(1)-Zn(1)-Cl(1)#1, Cl(1)-Zn(1)-S(1), Cl(1)#1-Zn(1)-S(1) bond angles are 112.85(2)°, 110.761(14)° and 106.083(13)°, respectively. These angles deviate from the regular tetrahedral value of 109.47°. These distortions results in secondary deviations of the other perinuclear angles from the ideal value. The deviation from ideal tetrahedral geometry is probably due to the steric hindrance of the 1,3-diisopropylthiourea ligands and the slight difference in electro negativities value of the S and Cl atoms respectively. The bond angles around the zinc atom are in the range 100.64(5)-112.85(2)°. The average C-S and C-N distances [1.7363(14) and 1.4020(18) Å] respectively of the diisopropyl thiourea molecules agree well with the CSD values [1.725(19) and 1.322 [/fig]
[fig] Figure 4: This diagram shows two-dimensional hydrogen bonding network perpendicular to the c axis. All the methyl hydrogen's are omitted for clarity. The hydrogen bonds are shown as dotted lines. [/fig]
[fig] Figure 5: Projection viewed of [ZnCl 2 (diptu) 2 ] down [100]. All the methyl hydrogen's are omitted for clarity. The hydrogen bonds are shown as dotted lines. [/fig]
[table] Table 1: Analytical data for metal complexes of diisopropylthiourea. [/table]
[table] Table 2: Summary of crystal data and structure refinement for [ZnCl 2 (diptu) 2 ]. [/table]
[table] Table 6: Minimum inhibition concentrations (MIC) in mg/mL of the metal complexes. [/table]
|
Informal Providers—Ground Realities in South Asian Association for Regional Cooperation Nations: Toward Better Cancer Primary Care: A Narrative Review
# Introduction
Early cancer diagnosis and treatment are major global health issues that disproportionately affect low-and middle-income countries (LMICs). South Asia is one such region with eight countries that form South Asian Association for Regional Cooperation (SAARC), Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka with low to medium Human Development Index values (except Maldives). SAARC countries have 25% of the world population and contribute to 25% of annual global cancer incidence and 16% of global cancer mortality. [bib_ref] Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for..., Sung [/bib_ref] These numbers are likely an underestimate because of limited cancer and death registries. Delays in diagnosis and cancer treatment contribute to a significantly higher proportion of cancer deaths as compared with high-income countries. 1 SAARC nations have a high-level Technical Committee on Health and Expert groups that are dedicated to improving outcomes in noncommunicable disease including cancer and maximizing human resources for health.Minimizing delay in diagnosis of symptomatic cancer is associated with better survival, better quality of life, and lesser cost of treatment. [bib_ref] Is increased time to diagnosis and treatment in symptomatic cancer associated with..., Neal [/bib_ref] The WHO strategy for Universal Health Coverage in SAARC has emphasized community-based primary health care as a core component. Reforming diverse health systems needs data on the nature, distribution, and quality of the primary health care workforce. The 2016 workshop of the World Organization of Family Doctors noted the substantial deficiencies of primary health care in SAARC nations, inadequate public systems, predominant private care, out-ofpocket payments, catastrophic health-related expenses leading to poverty, and reliance on alternative healers. [bib_ref] Primary healthcare policy implementation in South Asia, Van Weel [/bib_ref] As of 2014, the health expenditure in South Asia was 4% as compared with 10% of gross domestic product globally and 13% in the Organisation for Economic Cooperation and Development countries [bib_ref] Health care expenditure and health outcome nexus: New evidence from the SAARC-ASEAN..., Rahman [/bib_ref] [fig_ref] TABLE 1: Comparative Chart on per Capita GDP Expenditure in Health [/fig_ref]. The modest investment in health of , 4% in all SAARC nations is made ineffective with emphasis on hospital-based and specialist care that relies on robust state capacity. SAARC nations have poor state capacity as reflected in the WHO workforce document. [bib_ref] The primary health-care system in China, Li [/bib_ref] The number of medical doctors per 10,000 population ranges from 7.35 to 11 in the SAARC nations. The comparative figure for China is 22.2, that for the United States is 26.1, that for Germany is 44, and that for Sweden is 71.The scarcity of workforce and the urbanrural disparity are at the heart of delayed cancer diagnosis and are inadequately researched.
Two recent systematic reviews focused on delays and barriers to cancer care in low-income countries, including SAARC and the interventions performed to address some of these barriers. [bib_ref] Delays and barriers to cancer care in low-and middle-income countries: A systematic..., Brand [/bib_ref] [bib_ref] Interventions addressing barriers to delayed cancer diagnosis in low-and middle-income countries: A..., Qu [/bib_ref] The lack of access was due to poor health literacy, stigma, poverty, loss of wages, lack of transportation, and geographical limitations. Other causes were inaccurate diagnosis, false reassurance, poor coordination of care, lack of referrals, and financial barriers. The interventions targeting health care professionals addressing these barriers reported only two studies conducted in SAARC (in Sri Lanka and Bangladesh). [bib_ref] Interventions addressing barriers to delayed cancer diagnosis in low-and middle-income countries: A..., Qu [/bib_ref] The WHO guide to cancer early diagnosis mentions lack of access to primary care as the step 1 barrier. 11 None of the interventions reviewed access to primary health care. These systematic reviews highlighted the urgent need for studies that understand access to affordable and quality health care. [bib_ref] Interventions addressing barriers to delayed cancer diagnosis in low-and middle-income countries: A..., Qu [/bib_ref] Up to 75% of the primary care workforce in SAARC nations is informal. [bib_ref] What is the role of informal healthcare providers in developing countries? A..., Sudhinaraset [/bib_ref] This workforce fills the gap created by lack of formally trained health care providers. The density, diversity, mobility, availability, education levels, composition, absenteeism, quality of the informal cadre, as benchmarked against formal workforce, and cost to system have not been adequately addressed in the mainstream health care literature. [bib_ref] Two Indias: The structure of primary health care markets in rural Indian..., Das [/bib_ref] [bib_ref] Quality of primary care in low-income countries: Facts and economics, Das [/bib_ref] [bib_ref] The impact of recall periods on reported morbidity and health seeking behavior, Das [/bib_ref] [bib_ref] Money for nothing: The dire straits of medical practice in Delhi, Das [/bib_ref] [bib_ref] In urban and rural India, a standardized patient study showed low levels..., Das [/bib_ref] This narrative review aims to provide an overview of the supply side provider distribution: who the providers are, how they practice, variation in their quality, and the provider network that sustains this ecosystem.
# Methods
An initial scoping exercise of the published literature on informal providers in primary health care in SAARC nations suggested that the data were too heterogeneous and/or sparse to allow a systematic review of qualitative studies as proposed by Dixon-Woods 18 or a theoretical qualitative meta-synthesis as proposed by Sandelowski. [bib_ref] Qualitative metasynthesis: Issues and techniques, Sandelowski [/bib_ref] We, therefore, performed a narrative review of the published literature in the English language from January 2000 to December 2021 using Informal Health Care Provider/Informal Provider, Primary Health Care as MeSH terms in the PubMed, Google Scholar, and Cochrane database of systematic reviews. All articles with the term Informal Health Care Provider/Informal Provider mentioned in the title or abstract were reviewed. Additional databases included those of the World Bank, Center for Global Development, American Economic Review, Journal Storage, and Web of Science. A manual search of other relevant articles and citation lists from the primary articles was performed. Gray literature in the English language that included policy blogs and allied publications was included. The review focused on papers in the SAARC nations that recognize the outsized presence and CONTEXT Key Objective Who are informal health care providers, and why do people choose to seek care from them? How does their quality of care compare with prevailing standards, and can they contribute to the early diagnosis of cancer? Knowledge Generated More than 75% of health care across the rural landscape in South Asian Association for Regional Cooperation nations is provided by untrained informal providers. They are accessible and affordable but often have substandard quality when compared with the prevalent standards of formal health care. Relevance Although the training of increased numbers of formal health care workforce should be the priority, simultaneously acknowledging and upskilling informal health care providers may be a cost-effective and pragmatic bridge toward universal health coverage. Their network and social relationships with their local communities could be leveraged to create referral pathways for early cancer diagnosis. Health policy debates should consider involving informal providers in the care continuum and also ensure quality and accountability. Abbreviations: GDP, gross domestic product; PPP, purchasing power parity; USD, US dollars.
## Jco global oncology
indispensable nature of informal health care providers, measured quality of care, the quantity of care provided as compared with formal health care providers, provider effort, interventions that attempt to upskill them, policy debates that have attempted to regularize and or regulate them, and themes that question the ethics of their existence and acceptance. The duration of the first two decades of the 21st century was chosen as this coincided with the 2001 UN sustainable development goals and directives to SAARC nations and the explicit focus on primary health care as key toward Universal Health Coverage.
# Results
## Definitions and scope
Informal providers or informal health care providers (IPs) form a vital cog in the wheel of rural health care ecosystem in SAARC nations. A systematic review published in 2013 [bib_ref] What is the role of informal healthcare providers in developing countries? A..., Sudhinaraset [/bib_ref] noted the absence of typology around informal providers.
Since the informality of care is context-dependent, rigid criteria do not apply. A set of flexible criteria that include most informal providers entail those 1. Who have no formal institutional training 2. Who get paid by patients per transaction 3. Who do not have registration, regulation, or oversight by any institution.
It is important to note that these exclude community health workers who are trained and recognized by nongovernmental organizations and governments, for instance, the Accredited Social Health Activist woman workforce in India who are the first port of call for any health-related event.The World Bank data that measure community health workers at 0.514 per 1,000 in South Asia neglects IPs. Although human resources in health review for India published in 2011 [bib_ref] Human resources for health in India, Rao [/bib_ref] suggested that many of these IPs are registered medical practitioners, there is lack of evidence of any such registry across SAARC nations.
A list of workers that belong to this IP group is shown in [fig_ref] TABLE 2: Informal Providers List of Search Terms for Informal Alternative health practitioner Local... [/fig_ref].
The definition of an IP leaves them out of the ambit of conventional methods of measurement (registration, regulation, and oversight), leading to underestimation. Quality of care is inferred from crude descriptive statistics of referral, treatment, and success rates. The number of IPs is inferred from common disease subtypes, maternal and child health, reproductive health, malaria, and HIV/AIDS.
## Why choose an ip?
There is a lack of trained formal doctors and state capacity to fill the gap lags by years 6 [fig_ref] TABLE 3: Health Care Workforce in SAARC Versus OECD NOTE [/fig_ref]. Access is defined by the presence of trained doctors and not IPs. Universal health care hinges on the notion that patients prefer qualified public health care to informal private providers.The Medical Advice Quality and Availability in Rural India (MAQARI) [bib_ref] Two Indias: The structure of primary health care markets in rural Indian..., Das [/bib_ref] study showed that patients chose both formal trained public providers and informal private providers for the same illnesses. The study showed that the difference in quality between an IP and an average public trained provider (standardized scores measured on the basis of knowledge using clinical vignettes and adherence to casespecific checklists) is sometimes small. An IP from a relatively richer state in southern India (Tamil Nadu) is sometimes better than a trained provider in some of the poorer states in north/central India (Bihar). [bib_ref] Two Indias: The structure of primary health care markets in rural Indian..., Das [/bib_ref] Proximity to households in both rural and urban settings leads to cost savings by avoiding travel and loss of daily wages. [bib_ref] Regulating Tanzania's drug shops-Why do they break the rules, and does it..., Goodman [/bib_ref] Formal care may be 5-15 times more expensive as compared with informal care. [bib_ref] The use of formal and informal curative services in the management of..., Amin [/bib_ref] IPs provide drugs cheaper than pharmacies by repackaging drugs in smaller units. [bib_ref] Regulating Tanzania's drug shops-Why do they break the rules, and does it..., Goodman [/bib_ref] [bib_ref] Private local pharmacies in low-and middle-income countries: A review of interventions to..., Smith [/bib_ref] IPs belong to the locality they serve, which may build trust and accountability and may help address social and cultural barriers as they understand local customs and practices. IPs allow flexible charges, deferred payments, accommodating financial insecurity, and lack of insurance. The longevity of association sustains a stable relationship compared with attrition and absence common in public clinics.
## Measuring quality of care
The quality of care given by a provider varies in each transaction, cannot be generalized, is difficult to measure, and differs on the basis of the illness and provider training. The methods used for measuring quality are standardized patients, medical vignettes, and direct observation. Other reported dimensions of quality are based on technical aspects of adhering to existing national guidelines and referral pathways. This has limitations when applied to SAARC nations as guidelines and referral pathways are often absent, and when present, they are poorly disseminated and provider awareness and knowledge may be poor. [bib_ref] What is the role of informal healthcare providers in developing countries? A..., Sudhinaraset [/bib_ref] A systematic approach to adapt and contextualize international guidelines is underway in India. [bib_ref] Adapting clinical guidelines in India-A pragmatic approach, Mehndiratta [/bib_ref] This pragmatic approach allows a framework to alter guidelines to suit local needs, affordability, and availability of resources. [bib_ref] Adapting clinical guidelines in India-A pragmatic approach, Mehndiratta [/bib_ref] In a large country like India, there is a huge geographic variance in the competence of IPs. The accuracy of diagnosis in the best performing states like Tamil Nadu and Gujarat for eclampsia, tuberculosis, and dysentery was 94%, 93%, and 91%, respectively. The figures for the worst performing states (Jharkhand, Bihar, and Uttar Pradesh) were 12%, 5%, and 16%, respectively.
In a study performed in rural Madhya Pradesh, India, 67% of health care providers did not have any medical qualifications. [bib_ref] In urban and rural India, a standardized patient study showed low levels..., Das [/bib_ref] Even at public health facilities, 63% of the care was provided by staff who had no medical training. This was due to chronic understaffing and absenteeism among qualified staff. Studies performed in urban and rural India and a similar study from China had surprisingly similar results, despite the large difference in the gross national income per capita ($583 US dollars (USD) in Madhya Pradesh and $3,179 USD in Shaanxi, China). [bib_ref] Survey using incognito standardized patients shows poor quality care in China's rural..., Sylvia [/bib_ref] The average time spent per patient was , 2 minutes, , 20% of the necessary questions were asked for emergent and nonemergent needs, only half the patients had close to correct treatment, and less than one fourth had correct diagnoses.
## Provider effort
Studies have shown provider effort to be a critical factor for better quality care. The effort is measured as a know-do gap between what providers know and the measured facts of patient interaction (what they do). Competence is measured by standardized scores using medical vignettes and performance by observing them in practice and measuring adherence to checklists. Das and Hammer 14 in the MAQARI study hypothesize that a 45°slope of correlation between knowledge and effort would imply that providers do what they know. This is observed at the lowest levels of competence, but as knowledge and training gets better, a larger gap is seen between what one knows and what one does [fig_ref] FIG 1: The know-do gap in medical care [/fig_ref]. The gap is worse in the public sector, but remains large in the private sector too. Health policy frameworks attribute this to overburdening or lack of equipment or infrastructure.
Das et alin the 2013 study measured this phenomenon in rural Madhya Pradesh, India, among public practitioners, 83% of whom also had private clinics. The know-do gap was less in the private practice than in public clinics. The same providers spent more time, reached more accurate diagnoses, and aligned better treatment.This could imply effects of financial incentives on better effort in the private sector.
The administrative accountability that drives incentives in the public sector is seldom enforced. A strong incentive for poor patients is the availability of subsidized or free drugs in the public sector. This leads to a curious effect of over prescription for financial gain in the private sector and over prescription in lieu of care in the public sector. [bib_ref] Antibiotic overuse in the primary health care setting: A secondary data analysis..., Sulis [/bib_ref] Leonard et al have measured the role of nonfinancial incentives 32 in increasing motivation for better care. When the provider was aware that patient transactions were monitored for compliance and quality, an immediate high and a sustained improvement was noted (Hawthorne effect). These findings have been replicated across LMICs and high-income countries. Bjorkmann and Svenson used this effect in rural Uganda by employing community workers to improve local accountability of the public providers. [bib_ref] Power to the people: Evidence from a randomized field experiment on community-based..., Björkman [/bib_ref] A reduction in the under-five mortality (33% improvement one year into the program) was seen in the experimental villages without effort in improving training or infrastructure.
The experience from national tuberculosis control programs in India suggest that IPs who are trained perform well with triage, early referral, leading to better outcomes in patients with tuberculosis. [bib_ref] Role of informal healthcare providers in tuberculosis care in low-and middle-income countries:..., Thapa [/bib_ref] These findings may be applicable for improving early diagnosis of both symptomatic and asymptomatic cancers.
## Mapping the types and number of informal providers
A systematic mapping of the network of these providers is lacking. The MAQARI study has the largest nationwide data across 19 Indian states, which is representative of 90% of rural households. Few similar studies have been performed outside of India, notably in Ghana, [bib_ref] Smith F: Community pharmacy in Ghana: Enhancing the contribution to primary health..., Adu-Sarkodie [/bib_ref] Zambia, [bib_ref] Can biomedical and traditional health care providers work together? Zambian practitioners' experiences..., Kaboru [/bib_ref] Nigeria, [bib_ref] Interactions between patent medicine vendors and customers in urban and rural Nigeria, Brieger [/bib_ref] Bangladesh, [bib_ref] Non-professional health practitioners and referrals to facilities: Lessons from maternal care in..., Parkhurst [/bib_ref] [bib_ref] Informal sector providers in Bangladesh: How equipped are they to provide rational..., Ahmed [/bib_ref] and Nepal. [bib_ref] A traditional healers' training model in rural Nepal: Strengthening their roles in..., Poudyal [/bib_ref] A brief summary of the MAQARI study describes the landscape that is similar across SAARC. [bib_ref] Two Indias: The structure of primary health care markets in rural Indian..., Das [/bib_ref] The study notes that of the 1,519 Indian Villages surveyed, 74% had at least one health care provider and 86% of providers worked in the private sector. Of the 3,473 providers surveyed, 68% were IPs in the private sector, 24% (842) belong to the AYUSH cadre (Ayurveda, Unani, and Homoeopathy degrees; formal providers trained in alternative nonallopathic streams), and only 8% (264) had an MBBS. degree. The average village (range 1,000-5,000) had 5.4 IPs. The proportion of these providers did not change significantly in more developed states (as per the socioeconomic index). The systematic review by Sudhinaraset et al 12 from 2013 shows similar high rates of IPs across countries, 65%-77% in Bangladesh, 55%-77% in Thailand, 77% in Uganda, 36%-49% in Nigeria, and 33% in Kenya. The summary of the MAQARI study estimates this number at 80% for India.
## Low quality of care and diagnostic delay: causal links
The WHO guide to cancer early diagnosis framework states that access to care, evaluation of disease, and referral for subsequent care are the three pivotal steps to good cancer outcomes. Access to some form of health care is near universal in SAARC nations. It is the substandard quality of care that undermines the utility of this access. Tertiary care referral institutes see a huge proportion of patients with untreated or improperly treated cancer with disease at an advanced stage.
The key barriers to early diagnosis of cancer as shown in a systematic review 9 include health illiteracy; cancer stigma; lack of access to reliable primary care; inaccurate diagnoses; limited access to diagnostics; poor coordination of care; and geographical, financial, and sociocultural factors. This poor coordination of quality of care is the causal link that mediates inaccurate diagnoses and poor referral, leading to advanced stage at presentation, which worsens mortality, thereby also feeding into the stigma surrounding cancer. Of the more than 300 studies that reported delay and barriers to early cancer diagnosis, only 25 studied interventions. Furthermore, very few described the prominent role played by IPs.
The success of interventions addressing barriers to delayed diagnosis had 12 of 25 studies performed in groups of health care professionals, including doctors, nurses, midwives, dental professionals, and student health care professionals. However, none of these studies included IPs whose proportion outnumbers the formal health care worker by a factor of 3 or higher in most LMICs. 12
## Publicly funded primary health care and informal providers: policy implications
There is an implicit acceptance that IPs are part of the health care fabric of SAARC nations. A national sample Studies show that a majority of these are IPs. [bib_ref] In urban and rural India, a standardized patient study showed low levels..., Das [/bib_ref] [bib_ref] Where is the public health sector?": Public and private sector healthcare provision..., Costa [/bib_ref] [bib_ref] Where are healthcare providers? Exploring relationships between context and human resources for..., Costa [/bib_ref] [bib_ref] Informal rural healthcare providers in North and South India, Gautham [/bib_ref] Surveys in their current format are unable to distinguish formal and informal health care providers, and this potentially misleads policy makers. Acknowledgment and measurement of this workforce is essential. Under the Indian Medical Act 1956, it was illegal for anyone to practice medicine and dispense drugs without a university medical qualification. The current National Medical Commission Act of India makes provision for temporary licenses to community health providers for primary and preventive care. The exact nature of these regulations and scope of licenses are unclear. Any specter of regulation from powerful governmental agencies is construed as a threat to existence. The establishment worries that legitimizing them is akin to promoting illegal activity. As of today, enforcement of regulations and punitive actions is not feasible because of lacking state capacity and strong support from local communities and local politicians. Recognizing this, many Indian states under the federal health structure, including West Bengal, Andhra Pradesh, Bihar, and Chhattisgarh, sought ways to train these cadres as community health care workers or paramedics. [bib_ref] Can interventions improve health services from informal private providers in low and..., Shah [/bib_ref] IPs have participated in polio vaccination drives, tuberculosis treatment programs, UNICEF's social mobilization programs, mother and child health programs, and distribution of contraception. Das et al [bib_ref] Two Indias: The structure of primary health care markets in rural Indian..., Das [/bib_ref] They concluded that upskilling IPs may be an effective short-term strategy to complement the ailing health care system. Before this trial that focused on upskilling a broad range of tasks that are routinely performed by IPs in primary health care, randomized trials showed improved performance in specific tasks such as treatment of malaria, [bib_ref] Impact of ministry of health interventions on private medicine retailer knowledge and..., Abuya [/bib_ref] HIV-focused care study in Pakistan, [bib_ref] Can interventions improve health services from informal private providers in low and..., Shah [/bib_ref] care for sexually transmitted illnesses in Peru, [bib_ref] Training pharmacy workers in recognition, management, and prevention of STDs: District-randomized controlled..., Garcia [/bib_ref] and training pharmacists in correctly treating cases of urethral discharge. [bib_ref] Training pharmacy workers in recognition, management, and prevention of STDs: District-randomized controlled..., Garcia [/bib_ref] Das et al make a compelling argument that at an annual salary of $6,000 USD, the government could hire 11 MBBS doctors and the same amount could train 360 IPs yearly who were within 5 km of an average household. This annual salary is not reflected in the additional infrastructural, staffing, and pension-related costs that governments have to bear. A systematic scoping review in LMICs 34 noted the positive impact on outcomes of employing IPs after support and training in detection, diagnosis, and administering antituberculous therapy. A similar policy focusing on training IPs to detect signs of early cancer and encourage referral to the nearest known cancer center may improve outcomes. The high-level meetings of the SAARC Technical Committee on Health and Population activities could be the focal point for consideration and debate on this issue.
The continuous focus on funding and improving services at rural primary health care centers (PHCs) and HWCs is vital. The 2022 audit report conducted in 18 states in India noted increased satisfaction levels, reduced out-of-pocket expenditure, and a move from private unqualified healers to qualified HWCs.This is in line with the gradual replacement of traditional birth attendants with institutional health care delivery of pregnant women over the past two decades in LMICs. [bib_ref] Traditional birth attendants and birth outcomes in low-middle income countries: A review, Garces [/bib_ref] The ambitious national project to operationalize 200,000 HWCs in India to strengthen the rural primary health care infrastructure was expected to have labor deficits. The 2022 audit report also noted that most states have failed to employ the full range of workers at PHCs and HWCs.These include nurses, pharmacists, and ancillary health care workers. IPs could be incentivized to fill this gap and align with the goals of PHC within the formal health care ecosystem.
# Limitations
We recognize the limitations of this narrative review as studies were not objectively selected on the basis of methodological quality. However, the studies in this review had methodological rigor, large sample size, and standardof-care metrics for evaluation of qualitative and quantitative aspects of health care provision. The conclusions matched the real-world scenario, and estimates have remained robust over a decade. The demography of the populations and the shared cultures allows a safe extrapolation for a policy recommendation across SAARC. We recognize the social, political, and legal challenges of acknowledging informal providers. The system-wide changes that follow the act of acknowledgment are unclear. Regulating an emergent space runs the risk of destroying its fabric as it may create entry barriers and special interest groups. As the economy improves, capacity for producing expertise and demand for better care will increase and the space for informal providers will inevitably shrink.
There remain systemic concerns about the perverse financial motives and harmful, wasteful practice of some IPs. However, they continue to thrive albeit in smaller numbers in states and nations that have a high coverage and proficiency in rural primary health care (Tamil Nadu in India, Sri Lanka, sand Bangladesh). The WHO Advisory Group under the Health System and Governance department suggested aligning private for-profit and notfor-profit actors to the public sector. 54 Engagement of IPs may foster accountability and enable the rural population as the key stakeholder to bridge and maximize services provided by either sector. This is vital as the incentive structure, motivation, and moral high ground of those who commit to the public rural health care system are tenuous. It must be valued and preserved. The funding for the formal health care provider sector should be a higher priority and must not compete with the regulatory or upskilling needs of the IP sector. The key challenge for informed policy is lack of information and systematic research. Governments must sponsor studies to understand and measure the impact that IPs and other for profit systems have on rural health care practice and delivery. [bib_ref] The informal healthcare providers and universal health coverage in low and middle-income..., Kumah [/bib_ref]
# Discussion
In conclusion, SAARC nations have a plethora of informal providers forming the base of the health care worker pyramid. They provide options to people who have poor access to formal health care systems. They have deep local roots, strong social connections, and established networks of care that covers a majority of remote villages. This sector is emergent and unregulated and has a wide variation in the quality of care delivered. Available but poor care worsens health outcomes. Integrating these providers into the mainstream is controversial with no consensus. Strict regulation, which disallows IPs, may reduce some harmful care, but will leave a segment of the population with no access to health care. The state capacity for such regulation is lacking. Acknowledging the IPs and frequent training may improve equity, quality, and accountability of care. It will be complementary to parallel efforts to improve the number and quality of doctors, nurses, and other formal health care workers. Training IPs for general education, cancer screening, early identification of symptomatic cancers, prompt referral to subdistrict or district hospitals for workup, and sensitization about palliative care is an urgent unmet need. The Lancet Commission 56 for sustainable development goals for high-quality health systems outlined four values: that systems are for people, equitable, resilient, and efficient. The network of IPs may satisfy these values as building blocks. This network must complement and align with the needs of the public rural health care ecosystem, allowing the public system to hold them accountable.
[fig] FIG 1: The know-do gap in medical care. The horizontal axis plots what a provider knows, as measured by medical vignettes, using percentage compliance with a medically necessary case-specific checklist of history questions and examinations (by rotating the curve). The vertical axis plots what the provider actually did with a similar patient, observed in practice. Every history question and examination can be compared in a pairwise comparison. If providers did everything they told us they would do, we should observe them on the 45°line. At very low levels of knowledge, practice is constrained by knowledge; at higher levels of knowledge, there is a significant gap between knowledge and practice; the know-do gap is larger in the public sector (in which there is no correlation between practice and knowledge), but even in the private sector, there is a significant gap at higher levels of knowledge. Used with permission of Annual Reviews, Inc, from Annual Review of Economics, Quality of Primary Care in Low-Income Countries: Facts and Economics, Das et al, 6:525-553, 2014; permission conveyed through Copyright Clearance Center, Inc. 14 survey of India in 2014 showed that 70% of patients visited a private health care worker as the first point of contact. [/fig]
[table] TABLE 1: Comparative Chart on per Capita GDP Expenditure in Health [/table]
[table] TABLE 2: Informal Providers List of Search Terms for Informal Alternative health practitioner Local medical practitioner [/table]
[table] TABLE 3: Health Care Workforce in SAARC Versus OECD NOTE. The differences between SAARC and OECD nations in physician and nursing workforce per 1,000 population are stark and likely to be sustained at least until 2030. The 2013 figures are from the WHO global observatory, and the 2030 figures are forecasted.Abbreviations: OECD, Organisation for Economic Cooperation and Development; SAARC, South Asian Association for Regional Cooperation. [/table]
|
Neoadjuvant chemotherapy for epithelial ovarian cancer in Japan: a JSGO-JSOG joint study
The incidence of ovarian cancer is steadily increasing in Japan, with more than half of the women with ovarian cancer having advanced disease at presentation[1,2]. The standard treatment for epithelial ovarian cancer comprises primary debulking surgery (PDS), followed by platinum-based chemotherapy[3]. However, PDS is associated with a relatively high risk of morbidity and mortality because of its invasiveness. Previous studies demonstrated non-inferiority of overall survival in women with advanced ovarian cancer who received neoadjuvant chemotherapy followed by interval debulking surgery (NACT/IDS) compared with those who underwent PDS [4-6]. More recent approaches for the initial treatment of epithelial ovarian cancer may have changed. Therefore, we examined temporal trends in the utilization of NACT and changes in survival for epithelial ovarian cancer in Japan.We conducted a retrospective observational study using data from the Japan Society of Obstetrics and Gynecology (JSOG) gynecological tumor registry database from 2002 to 2015 (n=48,426). This database is an organ-based cancer registry of gynecological malignancies that is supported and managed by the gynecologic tumor committee of JSOG; the database records cancer subtypes, characteristics, treatments, and survival data[7]. The present study was conducted after obtaining approval from our Institutional Review Board.Information extracted from the database included patient demographics (age, area of registry, and calendar year at diagnosis), tumor histology (serous, mucinous, endometrioid, and clear cell), treatment type (staging surgery, including lymphadenectomy and NACT), and survival outcome (cause-specific survival (CSS). Patients were classified by age according to the definition of the Ministry of Health, Labor and Welfare in Japan: non-elderly (<65 years), young-elderly (65-74 years), and elderly (≥75 years)[8,9].The primary objective of this study was to examine time-specific trends of the proportion of women treated using NACT/IDS among patients with primary epithelial ovarian cancer in Japan. The secondary objective was to investigate the temporal trends of the 5-year CSS rate from 2002 to 2011.Statistical analyses were performed as follows. Continuous variables were assessed using the Mann-Whitney U test or Student's t-test. Ordinal and categorical variables were analyzed
using χ 2 -test. The Joinpoint Regression Program (version 4.7.0.0), provided by the National Cancer Institute (Bethesda, MD, USA), was utilized to evaluate temporal trends, which were analyzed using linear segmented regression. Log transformation was then performed to determine the annual percent change of the slope with 95% confidence intervals (CIs). All hypotheses were 2-tailed, and values of p<0.05 were considered to indicate statistical significance. Statistical analysis was performed using Statistical Package for Social Sciences (version 25.0; IBM SPSS, Armonk, NY, USA).
Among 48,426 women, 5,153 (10.6%, 95% CI=10.3-10.9) received NACT. Women who received NACT were more likely to be old, have a recent diagnosis, and be residents of East Japan, but were less likely to be residents of West Japan (p<0.05 for all; . During the study period, there was a significant increase in the utilization of NACT among the entire cohort: 7.2% to 14.9% (2.1-fold increase, p<0.001;.
Stratification by patient agerevealed a significant increase in NACT use among the 3 age groups, with the elderly group exhibiting the largest interval increase (non-elderly, .
The present study demonstrated a significant increase in the utilization of NACT in Japan, with one in 7 women with epithelial ovarian cancer receiving NACT in 2015. This increased use of NACT was particularly prominent in elderly women as well as those with clear cell histology. Our study also indicated cohort-level increases in the utilization of NACT as well as 5-year survival rates. A recent study in a US population also reported a shift toward the use of NACT that was associated with a decrease in postoperative mortality and improved cohort-level survival. Year . Temporal trends of the 5-years cause-specific survival rates (cohort level). Y-axis is truncated to 50%-100%. The 5-yeasrs cause-specific survival rate with ovarian cancer per calendar year is shown. Lines represent modeled estimates. Dots represent actual observed values and bars represent 95% confidence interval.
Patient selection for NACT/IDS should be personalized based on tumor histology because chemoresistant tumors such as clear cell carcinoma may be disadvantageous to this treatment strategy. Japanese women have a higher frequency of clear cell carcinoma occurrence compared with that seen in the US or other Western countries. Therefore, NACT/IDS may not always be a substitute for PDS for the initial treatment of epithelial ovarian cancer in Japanese women. Further studies are required to assess the association between NACT/IDS and this specific histology.
Our study has some limitations. This was a retrospective study; therefore, there may be confounding factors that could have affected the results. For instance, data for cancer stage prior to NACT, indications for NACT, and tumor differentiation to distinguish highgrade and low-grade serous tumors were not available in the JSOG database. Moreover, the database only focuses on leading hospitals in Japan, such as university hospitals and cancer centers, thereby increasing the possibility of selection bias.
In conclusion, there was a significant increase in the use of NACT for epithelial ovarian cancer in Japan. The shift toward NACT may be associated with improved survival among the entire cohort. While this trend may be encouraging for the treatment of chemosensitive tumors with high-grade serous ovarian cancer, its oncological safety remains undetermined in chemoresistant histology types. Therefore, consideration of NACT/IDS in women with ovarian cancer requires careful counseling and individualized treatment discussions with patients due to the unknown safety of this treatment among this population. |
Association of Trimethylamine N-Oxide and Related Metabolites in Plasma and Incident Type 2 Diabetes
et al. Association of trimethylamine N-oxide and related metabolites in plasma and incident type 2 diabetes: the Cardiovascular Health Study. JAMA Netw Open. 2021;4(8):e2122844.
## Etable 1. spearman correlation coefficients between baseline metabolite plasma
## Concentrations among 4442 participants in the cardiovascular health study
## Etable 4. association of serial measures of plasma tmao and related metabolites with incident type 2 diabetes among 4442 participants in the cardiovascular health study with follow-up restricted to the first 12 years
## -------multivariable model-----------multivariable
proportional-hazards regression analysis of incident type 2 diabetes that included the metabolite (linear variable), covariates and a multiplicative term between the metabolite (linear variable) and the effect modifier. Listed in the table are p-values for the multiplicative terms. The covariates included age, sex, race, site, education, income, BMI, waist, smoking, physical activity, systolic blood pressure, hypertension, LDL, CHD, animal sourced foods consumption and total energy intake. The pre-specified threshold of significance for these interaction tests was 0.0014 (0.05/36 tests). |
Advances in the Pathophysiology of Thrombosis in Antiphospholipid Syndrome: Molecular Mechanisms and Signaling through Lipid Rafts
Citation: Capozzi, A.; Manganelli, V.; Riitano, G.; Caissutti, D.; Longo, A.; Garofalo, T.; Sorice, M.; Misasi, R. Advances in the Pathophysiology of Thrombosis in Antiphospholipid Syndrome: Molecular Mechanisms and Signaling through Lipid Rafts. J.Clin. Med. 2023, 12, 891. https:// Abstract: The pathological features of antiphospholipid syndrome (APS) are related to the activity of circulating antiphospholipid antibodies (aPLs) associated with vascular thrombosis and obstetric complications. Indeed, aPLs are not only disease markers, but also play a determining pathogenetic role in APS and exert their effects through the activation of cells and coagulation factors and inflammatory mediators for the materialization of the thromboinflammatory pathogenetic mechanism. Cellular activation in APS necessarily involves the interaction of aPLs with target receptors on the cell membrane, capable of triggering the signal transduction pathway(s). This interaction occurs at specific microdomains of the cell plasma membrane called lipid rafts. In this review, we focus on the key role of lipid rafts as signaling platforms in the pathogenesis of APS, and propose this pathogenetic step as a strategic target of new therapies in order to improve classical anti-thrombotic approaches with "new" immunomodulatory drugs.
# Introduction
Antiphospholipid syndrome (APS) is a multisystemic disorder first described in the 1980s as an autoantibody-induced thrombophilia. The pathological features of APS are related to the activity of circulating antiphospholipid antibodies (aPLs) associated with vascular thrombosis and obstetric complications [bib_ref] The antiphospholipid syndrome, Levine [/bib_ref] [bib_ref] The antiphospholipid syndrome in patients with systemic lupus erythematosus, Pons-Estel [/bib_ref] [bib_ref] International consensus statement on an update of the classification criteria for definite..., Miyakis [/bib_ref]. APS usually occurs in the fourth decade of life, and patients can be classified into two groups: primary disease, if it presents in the absence of another autoimmune disorders, or secondary disease, if it is associated with a concomitant connective disease, most commonly systemic lupus erythematosus (SLE) [bib_ref] Epidemiology of the antiphospholipid antibody syndrome, Petri [/bib_ref]. In patients with APS, thrombosis can affect arterial, venous and microvascular circuits; however, arterial ones are more severe and more commonly found in male than female patients. They frequently occur in the cerebral circulation, causing stroke and transient ischemic attack [bib_ref] Antiphospholipid syndrome: Clinical and immunologic manifestations and patterns of disease expression in..., Cervera [/bib_ref].
Recurrent miscarriage is the main obstetric complication of obstetrical APS (OAPS); however, other events, such as unexplained fetal death or premature birth related to placental insufficiency, eclampsia or preeclampsia, may also represent adverse events associated with OAPS [bib_ref] Antiphospholipid syndrome: Clinical and immunologic manifestations and patterns of disease expression in..., Cervera [/bib_ref] [bib_ref] Pregnancy and antiphospholipid syndrome, Schreiber [/bib_ref].
Beyond vascular events and pregnancy morbidities, the clinical spectrum of the APS may include additional extra-criteria manifestations that cannot be explained solely by the onset of a prothrombotic state. The first description of APS by Hughes and colleagues already included the involvement of the nervous system [bib_ref] Thrombosis, abortion, cerebral disease, and the lupus anticoagulant, Hughes [/bib_ref]. The pathogenesis concerns not only thrombo-occlusive events, but refers to neurological manifestations, such as migraine, epilepsy, chorea, and cognitive dysfunctions, probably as a consequence of the direct interaction of aPLs with nervous tissues [bib_ref] Neurological manifestations of antiphospholipid syndrome, Rodrigues [/bib_ref] [bib_ref] Neurological features of the antiphospholipid (Hughes') syndrome, Hughes [/bib_ref]. The presence of aPLs is often associated with hematological manifestations, particularly in patients with idiopathic thrombocytopenic purpura (ITP), where aPLs may be responsible for the induction of thrombocytopenia [bib_ref] Antiphospholipid antibodies and antiphospholipid syndrome in patients presenting with immune thrombocytopenic purpura:..., Diz-Kucukkaya [/bib_ref] [bib_ref] The hematologic manifestations of the antiphospholipid syndrome, Uthman [/bib_ref]. Among the cutaneous manifestations of APS, the most common is livedo reticularis, but pseudovasculitic lesions, necrotic skin ulcers and digital gangrene can also be found [bib_ref] Antiphospholipid syndrome: Clinical and immunologic manifestations and patterns of disease expression in..., Cervera [/bib_ref] [bib_ref] Dermatologic manifestations of the antiphospholipid syndrome: Two hundred consecutive cases, Frances [/bib_ref] [bib_ref] Cutaneous manifestations of antiphospholipid syndrome: A review of the clinical features, diagnosis..., Pinto-Almeida [/bib_ref]. A minority of patients (less than 1%) may develop a severe variant of the disease, defined catastrophic APS (CAPS) and described as an acute multiorgan dysfunction, with a rapid onset of microvascular thrombosis in at least three organs; a mortality rate of more than 50% is reported in cases not promptly treated.
The antigens recognized by aPLs are phospholipids, phospholipid/protein complexes and phospholipid-binding proteins, among which β2-glycoprotein I (β2-GPI) is now recognized as the major autoantigen with the highest antibody-binding activity in APS patients [bib_ref] Beta2-glycoprotein I-dependent lupus anticoagulant highly correlates with thrombosis in the antiphospholipid syndrome, De Laat [/bib_ref].
The association between aPLs and thrombosis is now well documented, just as it is ascertained that these autoantibodies are not only markers of disease but are even responsible for the induction of a procoagulant phenotype and therefore for triggering the pathophysiology of APS. However, although several mechanisms have been proposed to describe the pathogenesis of APS, it is difficult task, due to the heterogeneity of autoantibody profiles and diversity in the potential effector functions of aPLs [bib_ref] Pathophysiology of the antiphospholipid antibody syndrome, Willis [/bib_ref] [bib_ref] Where does it come from?, Sherer [/bib_ref] [bib_ref] Mechanisms of thrombosis in systemic lupus erythematosus and antiphospholipid syndrome, De Groot [/bib_ref].
In this review, we describe the pathogenic mechanisms of APS through the key role of lipid rafts as signaling platforms, suggesting this pathogenetic step as a strategic target of new therapies.
## Mechanisms in the pathophysiology of aps
Evidence of the thrombogenic activity of aPLs has been provided by both in vitro and in vivo studies; however, despite the presence of aPLs in the circulation, thrombotic manifestation is not always observed in the patients [bib_ref] From antibody to clinical phenotype, the black box of the antiphospholipid syndrome:..., Du [/bib_ref]. Moreover, several results in experimental animal models demonstrate that the infusion of aPLs alone does not cause spontaneous thrombotic complications; on the contrary, the appearance of a thromboinflammatory state can be observed after small injuries to the vessel wall or infusion of lipopolysaccharides and other immunostimulants which lead to a disruption of the vascular system. These considerations suggest the "two-hit hypothesis" concept to better clarify the pathogenetic role of aPLs in the mechanisms underlying the development of thrombosis in APS patients. According to the "two-hit hypothesis", the aPLs provide the first hit, inducing a thrombophilic state, but a second condition (second hit) such as trauma, surgery, infection, or other inflammatory stimulus is required for thrombosis to take place [bib_ref] In vivo distribution of β2 glycoprotein I under various pathophysiologic conditions, Agostinis [/bib_ref] [bib_ref] Pathogenesis of antiphospholipid syndrome: Understanding the antibodies, Meroni [/bib_ref] [bib_ref] Anti-β2-glycoprotein I, antiprothrombin antibodies, and the risk of thrombosis in the antiphospholipid..., Galli [/bib_ref] [bib_ref] β2-Glycoprotein-1 autoantibodies from patients with antiphospholipid syndrome are sufficient to potentiate arterial..., Arad [/bib_ref]. Thrombus formation is the most studied aspect of APS pathophysiology, and the procoagulant phenotype observed in the disease has been described as a result of the synergistic activation of many elements, where inflammation represents a central pathogenetic factor that acts as a mediator between the coagulation dysfunction and thrombotic manifestations. Several studies have shown that aPLs exert their effects through activation of various cells (endothelial cells, monocytes, platelets, endometrial and decidual cells) [bib_ref] The Journey of Antiphospholipid Antibodies from Cellular Activation to Antiphospholipid Syndrome, Willis [/bib_ref] , as well as a complement system, coagulation factors and inflammatory mediators [bib_ref] Thrombus formation induced by antibodies to β2-glycoprotein I is complement dependent and..., Fischetti [/bib_ref]. These pathogenetic mechanisms underlying clinical manifestations of APS can be exploited as targets of immunomodulatory therapeutic strategies to be used in APS patients, as summarized in [fig_ref] Figure 1: Figure 1 [/fig_ref].
aPLs activate endothelial cells to release several proinflammatory cytokines and to express adhesion molecules, involving toll-like receptors (TLRs) and LDL receptor related protein 8 (LRP8) [bib_ref] Endothelial cells as target for antiphospholipid antibodies. Human polyclonal and monoclonal anti-beta..., Del Papa [/bib_ref] [bib_ref] Endothelial activation by aPL: A potential pathogenetic mechanism for the clinical manifestations..., Meroni [/bib_ref] [bib_ref] Role of the MyD88 transduction signaling pathway in endothelial activation by antiphospholipid..., Raschi [/bib_ref]. Moreover, anti-β2-GPI antibodies may activate monocytes via several signaling pathway kinases and through engagement of various TLRs, leading to a proinflammatory phenotype [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref] [bib_ref] Anti-beta(2)GPI/beta(2)GPI induced TF and TNF-alpha expression in monocytes involving both TLR4/MyD88 and..., Xie [/bib_ref]. Pathogenic aPLs also induce platelet activation and aggregation, where TLRs, LRP8 and the GPIbα subunit of the GPIb-V-IX play a role as candidate receptors responsible for signaling pathway activation [bib_ref] Platelet adhesion to dimeric beta-glycoprotein I under conditions of flow is mediated..., Pennings [/bib_ref] [bib_ref] Increased circulating platelet-leucocyte complexes and platelet activation in patients with antiphospholipid syndrome,..., Joseph [/bib_ref]. Pathogenetic mechanisms contributing to thrombosis in APS and possible molecular targets. Molecular mechanisms of aPLs, in particular of anti-β2-GPI antibodies, involve the cooperation of endothelial cells, monocytes and platelets with a production of adhesion molecules, proinflammatory cytokines and tissue factor, leading to a proinflammatory and procoagulant state. Anti-β2-GPI antibodies may also activate neutrophils by inducing the formation of NETs important to propagate inflammation and contribute to thrombosis. Inflammation and complement activation play a central role in upregulating cell activation in the pathophysiology of APS. These pathogenetic pathways underlying clinical manifestations of APS can be exploited as targets of immunomodulatory therapeutic strategies to be used in APS patients. Beta2-glycoprotein I (β2-GPI); Annexin 2 (ANXA2); toll-like receptors (TLRs); C3a receptor (C3a-R); C5a receptor (C5a-R); cholesterol (Chol); endothelial protein C receptor (EPCR); lyso-bis-phosphatidic acid (LBPA); myeloid differentiation factor 88 (MyD88); mammalian target of rapamycin complex (mTOR); nuclear factor-kappa B (NF-κB); neutrophil extracellular traps (NETs); vascular cellular adhesion molecule (VCAM-1); intercellular adhesion molecule-1 (ICAM-1); tumor necrosis factor alpha (TNFα); tissue factor (TF); interleukin (IL). aPLs activate endothelial cells to release several proinflammatory cytokines and to express adhesion molecules, involving toll-like receptors (TLRs) and LDL receptor related protein 8 (LRP8) [bib_ref] Endothelial cells as target for antiphospholipid antibodies. Human polyclonal and monoclonal anti-beta..., Del Papa [/bib_ref] [bib_ref] Endothelial activation by aPL: A potential pathogenetic mechanism for the clinical manifestations..., Meroni [/bib_ref] [bib_ref] Role of the MyD88 transduction signaling pathway in endothelial activation by antiphospholipid..., Raschi [/bib_ref]. Moreover, anti-β2-GPI antibodies may activate monocytes via several signaling pathway kinases and through engagement of various TLRs, leading to a proinflammatory phenotype [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref] [bib_ref] Anti-beta(2)GPI/beta(2)GPI induced TF and TNF-alpha expression in monocytes involving both TLR4/MyD88 and..., Xie [/bib_ref]. Pathogenic aPLs also induce platelet activation and aggregation, where TLRs, LRP8 and the GPIbα subunit of the GPIb-V-IX play a role as candidate receptors responsible for signaling pathway activation [bib_ref] Platelet adhesion to dimeric beta-glycoprotein I under conditions of flow is mediated..., Pennings [/bib_ref] [bib_ref] Increased circulating platelet-leucocyte complexes and platelet activation in patients with antiphospholipid syndrome,..., Joseph [/bib_ref].
A central player in the development of antibody-mediated thrombosis in APS is the tissue factor (TF), a primary cellular initiator of the extrinsic coagulation cascade. An upregulation of TF in endothelial cells, monocytes and platelets following anti-β2-GPI antibodies treatment has been demonstrated [bib_ref] Thrombosis in primary antiphospholipid syndrome: A pivotal role for monocyte tissue factor..., Cuadrado [/bib_ref] [bib_ref] Increased levels of tissue factor mRNA in mononuclear blood cells of patients..., Dobado-Berrios [/bib_ref] [bib_ref] Antiphospholipid antibodies from patients with the antiphospholipid syndrome induce monocyte tissue factor..., Lopez-Pedrera [/bib_ref]. Furthermore, the results of our paper showed that the expression of TF was significantly increased in platelets of patients with APS compared to those of healthy donors [bib_ref] Tissue factor over-expression in platelets of patients with anti-phospholipid syndrome: Induction role..., Capozzi [/bib_ref].
aPLs may induce the formation of neutrophil extracellular traps (NETs), which are mostly known to trap pathogens, but they also play a prothrombotic role activating platelets and clotting factors. It was highlighted that monoclonal anti-β2-GPI antibodies were able to promote NETs release in vitro [bib_ref] Mechanisms of immunothrombosis and vasculopathy in antiphospholipid syndrome, Knight [/bib_ref] [bib_ref] Anti-β2GPI/β2GPI induces neutrophil extracellular traps formation to promote thrombogenesis via the TLR4/MyD88/MAPKs..., Zha [/bib_ref]. In further APS case studies, increased Molecular mechanisms of aPLs, in particular of anti-β2-GPI antibodies, involve the cooperation of endothelial cells, monocytes and platelets with a production of adhesion molecules, proinflammatory cytokines and tissue factor, leading to a proinflammatory and procoagulant state. Anti-β2-GPI antibodies may also activate neutrophils by inducing the formation of NETs important to propagate inflammation and contribute to thrombosis. Inflammation and complement activation play a central role in upregulating cell activation in the pathophysiology of APS. These pathogenetic pathways underlying clinical manifestations of APS can be exploited as targets of immunomodulatory therapeutic strategies to be used in APS patients. Beta2-glycoprotein I (β2-GPI); Annexin 2 (ANXA2); toll-like receptors (TLRs); C3a receptor (C3a-R); C5a receptor (C5a-R); cholesterol (Chol); endothelial protein C receptor (EPCR); lyso-bis-phosphatidic acid (LBPA); myeloid differentiation factor 88 (MyD88); mammalian target of rapamycin complex (mTOR); nuclear factor-kappa B (NF-κB); neutrophil extracellular traps (NETs); vascular cellular adhesion molecule (VCAM-1); intercellular adhesion molecule-1 (ICAM-1); tumor necrosis factor alpha (TNF-α); tissue factor (TF); interleukin (IL).
A central player in the development of antibody-mediated thrombosis in APS is the tissue factor (TF), a primary cellular initiator of the extrinsic coagulation cascade. An up-regulation of TF in endothelial cells, monocytes and platelets following anti-β2-GPI antibodies treatment has been demonstrated [bib_ref] Thrombosis in primary antiphospholipid syndrome: A pivotal role for monocyte tissue factor..., Cuadrado [/bib_ref] [bib_ref] Increased levels of tissue factor mRNA in mononuclear blood cells of patients..., Dobado-Berrios [/bib_ref] [bib_ref] Antiphospholipid antibodies from patients with the antiphospholipid syndrome induce monocyte tissue factor..., Lopez-Pedrera [/bib_ref]. Furthermore, the results of our paper showed that the expression of TF was significantly increased in platelets of patients with APS compared to those of healthy donors [bib_ref] Tissue factor over-expression in platelets of patients with anti-phospholipid syndrome: Induction role..., Capozzi [/bib_ref].
aPLs may induce the formation of neutrophil extracellular traps (NETs), which are mostly known to trap pathogens, but they also play a prothrombotic role activating platelets and clotting factors. It was highlighted that monoclonal anti-β2-GPI antibodies were able to promote NETs release in vitro [bib_ref] Mechanisms of immunothrombosis and vasculopathy in antiphospholipid syndrome, Knight [/bib_ref] [bib_ref] Anti-β2GPI/β2GPI induces neutrophil extracellular traps formation to promote thrombogenesis via the TLR4/MyD88/MAPKs..., Zha [/bib_ref]. In further APS case studies, increased release of NETs in the neutrophils was demonstrated, and serum samples have high levels of circulating NETs as well as being defective in degrading NETs. Recently, anti-NET antibodies have been found in sera from patients with primary APS [bib_ref] Degradation of neutrophil extra-cellular traps is decreased in patients with antiphospholipid syndrome, Leffler [/bib_ref] [bib_ref] Anti-neutrophil extracellular trap anti-bodies and impaired neutrophil extracellular trap degradation in antiphospholipid..., Zuo [/bib_ref].
In addition, the thrombogenic effect of aPLs may involve the activation of the complement system, through both the alternative and the classical pathway. Experiments in mice showed that C3 and C5 activation was necessary for aPL-induced thrombosis [bib_ref] Complement activation in patients with isolated antiphospholipid antibodies or primary antiphospholipid syndrome, Breen [/bib_ref] [bib_ref] Requirement of activation of complement C3 and C5 for anti-phospholipid antibody-mediated thrombophilia, Pierangeli [/bib_ref].
There is evidence that C5a factor binds to endothelial cells, resulting in increased neutrophil adhesion, TF expression, and release of other procoagulant molecules [bib_ref] Complement C3 activation is required for antiphospholipid antibody-induced fetal loss, Holers [/bib_ref] [bib_ref] Complement C5a receptors and neutrophils mediate fetal injury in the antiphospholipid syndrome, Girardi [/bib_ref] [bib_ref] Complement activation in antiphospholipid syndrome and its inhibition to prevent rethrombosis after..., Meroni [/bib_ref].
Many studies have described the direct activity of aPLs in association with pregnancy complications as part of the pathogenesis of OAPS. In particular, aPLs may interact with trophoblast cells and consequently apoptosis, abnormal cell differentiation and decreased secretion of human chorionic gonadotropin may be observed. Inflammation plays a central role in inducing pregnancy morbidity in OAPS patients; in fact, at the utero-placental interface, aPLs induce a proinflammatory phenotype in the vasculature. These molecular mechanisms lead to endothelial injury and activation of monocytes and neutrophils involving the activation of the complement system [bib_ref] Modulation of trophoblast angiogenic factor secretion by antiphospholipid antibodies is not reversed..., Carroll [/bib_ref] [bib_ref] ApoE Receptor 2 Mediation of Trophoblast Dysfunction and Pregnancy Complications Induced by..., Ulrich [/bib_ref] [bib_ref] Antiphospholipid antibodies limit trophoblast migration by reducing IL-6 production and STAT3 activity, Mulla [/bib_ref].
The heterogeneity of the mechanisms underlying APS may be explained by the different aPL profiles, which also include extra-antibodies beyond those routinely tested [bib_ref] Closing the Serological Gap in the Antiphospholipid Syndrome: The Value of "Non-criteria"..., Zohoury [/bib_ref] [bib_ref] Non-criteria antiphospholipid antibodies": Bridging the gap between seropositive and seronegative Antiphospholipid Syndrome, Truglia [/bib_ref] ; thus, the clinical characteristics of the disease in an individual are due to different autoantibody specificities, together with genetic comorbidities and acquired risk factors.
At the end, the heterogeneity of the mechanisms underlying APS may be related to the different signal transduction pathways triggered by aPLs. Recently, increasing evidence suggests that they are driven by specific microdomains on the cell plasma membrane termed lipid rafts.
## Lipid rafts in the immune signaling
Cellular membranes are not homogenous mixtures of lipids and proteins, but some, such as free cholesterol and glycosphingolipids segregate into lipid rafts [bib_ref] Functional rafts in cell membranes, Simons [/bib_ref] [bib_ref] Sphingolipid-cholesterol rafts diffuse as small entities in the plasma membrane of mammalian..., Pralle [/bib_ref]. Since several investigations found that glycosphingolipid clusters are resistant to detergent extraction, these specialized subdomains have also designed detergent-resistant membranes (DRMs). In any case, lipid rafts are now universally defined as small (10-200 nm) heterogeneous membrane domains enriched in glycosphingolipid and cholesterol that regulate cellular polarity and vesicular traffic as well as cell signaling pathways [bib_ref] Sphingolipid-cholesterol rafts diffuse as small entities in the plasma membrane of mammalian..., Pralle [/bib_ref] [bib_ref] The ins and outs of lipid rafts: Functions in intracellular cholesterol homeostasis,..., Ouweneel [/bib_ref].
These microdomains are well known for their role in receptor signaling on the plasma membrane and are essential to cellular functions such as signal transduction and spatial organization of the cellular membrane. As a result, the structural properties of microdomains can promote compartmentalization of plasma membrane proteins that have a higher affinity for the liquid-ordered phase, while they will exclude proteins that have higher affinity for the liquid-disordered phase. In this scenario, some protein-protein interactions will be promoted, while others will be prevented. A complex network of lipid-protein, lipidlipid, and protein-protein interactions contribute to the activation of a variety of signal transduction pathways implicated in several processes, including apoptosis, proliferation, inflammation and autophagy [bib_ref] Dual acylation and lipid raft association of Src-family protein tyrosine kinases are..., Zaman [/bib_ref] [bib_ref] LFA-1 and PI3K but not SHP-2 interact with GM1-or GM3-enriched microdomains in..., Barbat [/bib_ref] [bib_ref] The death-inducing signalling complex is recruited to lipid rafts in Fas-induced apoptosis, Scheel-Toellner [/bib_ref] [bib_ref] Evidence for the involvement of GD3 ganglioside in autophagosome formation and maturation, Matarrese [/bib_ref] [bib_ref] On the role of sphingolipids in cell survival and death, Iessi [/bib_ref].
Two common types of lipid rafts have been described: planar lipid rafts, and invaginated rafts which curve inward, called caveolae. Caveolae are specialized lipid rafts found in very high numbers in adipocytes, which are formed by the polymerization of caveolin1 (CAV1) to generate concave regions on the cell membrane. Caveolae are involved in endocytosis of different proteins and can also play a role in signal transduction, although they may not be essential, since several cell types, such as neurons and lymphocytes, lack caveolin [bib_ref] On the role of sphingolipids in cell survival and death, Iessi [/bib_ref] [bib_ref] Caveolins, a family of scaffolding proteins for organizing "preassembled signaling complexes" at..., Okamoto [/bib_ref].
The formation of lipid rafts is not merely confined to the plasma membrane; indeed, similar lipid microdomains, more specifically known as lipid "raft-like microdomains", are distributed on the membrane of subcellular organelles, which include Golgi apparatus, nuclei, endoplasmic reticulum (ER) and mitochondria [bib_ref] Dissecting the role of the golgi complex and lipid rafts in biosynthetic..., Heino [/bib_ref] [bib_ref] Lipid microdomains in cell nucleus, Cascianelli [/bib_ref] [bib_ref] Endosomal compartment contributes to the propagation of CD95/Fas-mediated signals in type II..., Matarrese [/bib_ref] [bib_ref] Alteration of endoplasmic reticulum lipid rafts contributes to lipotoxicity in pancreatic β-cells, Boslem [/bib_ref] [bib_ref] Lipid microdomains contribute to apoptosis-associated modifications of mitochondria in T cells, Garofalo [/bib_ref]. Several studies have shown that these organelles differ both quantitatively and qualitatively in their lipid content. Raftlike microdomains are enriched in gangliosides and cholesterol, but with a lower content compared to plasma membranes. In addition, some components of raft-like microdomains are present within ER mitochondria-associated membranes [bib_ref] Mitochondria-associated ER membranes (MAMs) and lysosomal storage diseases, Annunziata [/bib_ref]. At these sites, key reactions can be catalyzed, with a significant impact on the regulation of intracellular trafficking, sorting and cell fate [bib_ref] Mitochondria-associated ER membranes (MAMs) and lysosomal storage diseases, Annunziata [/bib_ref] [bib_ref] Evidence for the involvement of lipid rafts localized at the ER-mitochondria associated..., Garofalo [/bib_ref] [bib_ref] Raft-like lipid microdomains drive autophagy initiation via AMBRA1-ERLIN1 molecular association within MAMs, Manganelli [/bib_ref].
It is well known that lipid rafts have been associated with several cell functions, including cell death. In fact, it has been suggested that lipid rafts can play a key role in the receptor-mediated apoptosis of T cells. Following CD95/Fas triggering, as well as tumor necrosis factor-family receptors (TNFRs), procaspase-8 oligomerization drives its activation through self-cleavage, activating downstream effector caspases and leading to apoptosis. Thus, activation of Fas results in receptor aggregation and formation of the so-called "deathinducing signaling complex" (DISC) [bib_ref] The death-inducing signalling complex is recruited to lipid rafts in Fas-induced apoptosis, Scheel-Toellner [/bib_ref] [bib_ref] An essential role for membrane rafts in the initiation of Fas/CD95-triggered cell..., Hueber [/bib_ref] , containing trimerized Fas, Fas-associated death domain (FADD) and procaspase-8 in lipid rafts, and also causing the recruitment of specific proapoptotic Bcl-2 family proteins to mitochondrial "raft like microdomains". The importance of lipid rafts in Fas-mediated apoptosis was further supported by the finding that the expression of membrane sphingomyelin, a major component of lipid rafts, enhances Fas-mediated apoptosis through increasing DISC formation, the activation of caspases, the efficient translocation of Fas into lipid rafts, and subsequently Fas clustering [bib_ref] Role of membrane sphingomyelin and ceramide in platform formation for Fas-mediated apoptosis, Miyaji [/bib_ref].
Whether changes in the composition or structure of lipid rafts play a role in autoimmunity is an important question which has started to be addressed in the last few years.
In particular, a role for lipid rafts in T cells activation was reported. In fact, the initial events of T-cell activation involve the movement of T-cell receptor (TCR) into lipid rafts [bib_ref] The role of lipid rafts in T cell antigen receptor (TCR) signalling, Janes [/bib_ref]. Peripheral blood T cells isolated from SLE patients were found to have more cholesterol and GM1 content in their plasma membrane compared with healthy individuals. This suggests an activated state of T cells in SLE, in which aggregated lipid rafts on the T-cell surface contain TCR as well as other co-stimulatory molecules [bib_ref] Altered lipid raft-associated signaling and ganglioside expression in T lymphocytes from patients..., Jury [/bib_ref]. Deng et al. reported that disruption of lipid rafts by methyl-beta-cyclodextrin (MβCD) delayed disease progression, whereas aggregation of rafts using cholera toxin accelerated disease progression in mice [bib_ref] Cholera toxin B accelerates disease progression in lupus-prone mice by promoting lipid..., Deng [/bib_ref].
In the APS, further studies demonstrated that anti-β2-GPI react with their target antigen, such as β2-GPI, annexin A2 (ANXA2), TLR2 and TLR4 within lipid rafts located in the plasma membrane of monocytes or endothelial cells, thereby producing a proinflammatory, procoagulant phenotype characterized by the release of TNFα and TF, respectively [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref] [bib_ref] New insights in glycosphingolipid function: "glycosignaling domain", a cell surface assembly of..., Hakomori [/bib_ref] [bib_ref] Endothelial cell activation by antiphospholipid antibodies, Meroni [/bib_ref] [bib_ref] Molecular Mechanisms of "Antiphospholipid Antibodies" and Their Paradoxical Role in the Pathogenesis..., Misasi [/bib_ref]. In [fig_ref] Figure 2: Figure 2 [/fig_ref] , the major intracellular signaling pathways triggered by anti-β2-GPI antibodies through lipid rafts and implicated in the procoagulant cell phenotype are depicted. In this regard, the first indication derived from the observation of Sorice et al. [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref] is that the anti-β2-GPI target antigen was found within lipid rafts which are specialized portions of the cell plasma membrane implied in signal transduction pathways, as revealed by the sucrose gradient analysis and coimmunoprecipitation experiments.
Interestingly, inappropriate changes in the size and/or structure of lipid rafts could influence their stability and may result in abnormal signaling. Signaling pathways induced by anti-β2-GPI antibodies. Scheme depicting the major intracellular signaling pathways triggered by anti-β2-GPI antibodies via TLRs and/or LRP6 through lipid rafts (dotted circle), implicated in the procoagulant endothelial cell phenotype characterized by the expression of TF. Lipid raft-targeted drugs act on cholesterol either as cholesterol depletors (MβCDs) or cholesterol synthesis inhibitors (statins). Annexin A2 (ANXA2); anti-β2-glycoprotein I (anti-β2-GPI antibodies); beta-catenin (β-catenin); cholesterol (Chol); interleukin-1 receptor-associated kinases (IRAKs); LDL receptor related protein 6 (LRP6); methyl-β-cyclodextrin (MβCD); myeloid differentiation factor 88 (MyD88); nuclear factor-kappa B (NF-κB); protease activated receptor-2 (PAR-2); tissue factor (TF); toll-like receptors (TLRs); TNF receptor associated factor 6 (TRAF6).
Interestingly, inappropriate changes in the size and/or structure of lipid rafts could influence their stability and may result in abnormal signaling.
## Signal transduction pathway through lipid rafts in the pathophysiology of aps
Signal transduction pathway(s) through lipid rafts play a key role in the proinflammatory and procoagulant events during APS, involving three cell types: endothelial cells, monocytes and platelets. Raschi et al. [bib_ref] Role of the MyD88 transduction signaling pathway in endothelial activation by antiphospholipid..., Raschi [/bib_ref] first demonstrated that anti-β2-GPI antibodies activate the TLR4 transduction signaling pathway in human endothelial cells, as revealed by transiently co-transfecting microvascular endothelial cells (HMEC-1), with dominantnegative constructs of different components of the pathway. Furthermore, the activation of these cells has been demonstrated by the phosphorylation of the myeloid differentiation factor 88 (MyD88), IRAK and NF-κB. Several receptors have been suggested to mediate β2-GPI/endothelial cell binding, including mainly ANXA2 and LRP8 [bib_ref] Mechanisms of Cellular Activation in the Antiphospholipid Syndrome, Müller-Calleja [/bib_ref]. Engagement of TLR4 activates both MyD88-dependent and MyD88-independent pathways, which may lead to the activation downstream of the MAP kinases. These pioneer studies on the signal transduction pathway triggered by anti-β2-GPI antibodies failed to demonstrate the role of lipid rafts. However, since it is known that plasma membrane TLRs are highly enriched in lipid rafts [bib_ref] Lateral diffusion of Toll-like receptors reveals that they are transiently confined within..., Triantafilou [/bib_ref] , the involvement of these domains in this pathway was further clarified in human monocytes [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref]. In this regard, it was noted that anti-β2-GPI antibody binding Signaling pathways induced by anti-β2-GPI antibodies. Scheme depicting the major intracellular signaling pathways triggered by anti-β2-GPI antibodies via TLRs and/or LRP6 through lipid rafts (dotted circle), implicated in the procoagulant endothelial cell phenotype characterized by the expression of TF. Lipid raft-targeted drugs act on cholesterol either as cholesterol depletors (MβCDs) or cholesterol synthesis inhibitors (statins). Annexin A2 (ANXA2); anti-β2-glycoprotein I (anti-β2-GPI antibodies); beta-catenin (β-catenin); cholesterol (Chol); interleukin-1 receptor-associated kinases (IRAKs); LDL receptor related protein 6 (LRP6); methyl-β-cyclodextrin (MβCD); myeloid differentiation factor 88 (MyD88); nuclear factor-kappa B (NF-κB); protease activated receptor-2 (PAR-2); tissue factor (TF); toll-like receptors (TLRs); TNF receptor associated factor 6 (TRAF6).
## Signal transduction pathway through lipid rafts in the pathophysiology of aps
Signal transduction pathway(s) through lipid rafts play a key role in the proinflammatory and procoagulant events during APS, involving three cell types: endothelial cells, monocytes and platelets. Raschi et al. [bib_ref] Role of the MyD88 transduction signaling pathway in endothelial activation by antiphospholipid..., Raschi [/bib_ref] first demonstrated that anti-β2-GPI antibodies activate the TLR4 transduction signaling pathway in human endothelial cells, as revealed by transiently co-transfecting microvascular endothelial cells (HMEC-1), with dominantnegative constructs of different components of the pathway. Furthermore, the activation of these cells has been demonstrated by the phosphorylation of the myeloid differentiation factor 88 (MyD88), IRAK and NF-κB. Several receptors have been suggested to mediate β2-GPI/endothelial cell binding, including mainly ANXA2 and LRP8 [bib_ref] Mechanisms of Cellular Activation in the Antiphospholipid Syndrome, Müller-Calleja [/bib_ref]. Engagement of TLR4 activates both MyD88-dependent and MyD88-independent pathways, which may lead to the activation downstream of the MAP kinases. These pioneer studies on the signal transduction pathway triggered by anti-β2-GPI antibodies failed to demonstrate the role of lipid rafts. However, since it is known that plasma membrane TLRs are highly enriched in lipid rafts [bib_ref] Lateral diffusion of Toll-like receptors reveals that they are transiently confined within..., Triantafilou [/bib_ref] , the involvement of these domains in this pathway was further clarified in human monocytes [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref]. In this regard, it was noted that anti-β2-GPI antibody binding on the surface of monocytic cells occurs within lipid rafts [bib_ref] Lipid rafts as a membrane-organizing principle, Lingwood [/bib_ref]. Biochemical analyses showed that the dimeric form of β2-GPI was present within lipid rafts [bib_ref] Anti-beta2-glycoprotein I antibodies induce monocyte release of tumor necrosis factor alpha and..., Sorice [/bib_ref] , which may be a consequence of an oxidation process [bib_ref] Oxidized beta2-glycoprotein I induces human dendritic cell maturation and promotes a T..., Buttari [/bib_ref]. Thus, β2-GPI interacts with lipid rafts only after dimerization, probably as a consequence of conformational changes, suggesting that oxidized β2-GPI is able to trigger signal transduction pathways [bib_ref] Oxidized beta2-glycoprotein I induces human dendritic cell maturation and promotes a T..., Buttari [/bib_ref] [bib_ref] Dimers of β2-glycoprotein I increase platelet deposition to collagen via interaction with..., Lutters [/bib_ref] and that β2-GPI dimers mimic in vitro the effects of β2-GPI-anti-β2-GPI antibody complexes [bib_ref] Dimers of β2-glycoprotein I increase platelet deposition to collagen via interaction with..., Lutters [/bib_ref]. In addition, it was found that ANXA2, the main receptor for β2-GPI [bib_ref] Cross-reactivity between annexin A2 and Beta-2-glycoprotein I in animal models of antiphospholipidsyndrome, Weiss [/bib_ref] [bib_ref] High affinity binding of β2-glycoprotein I to human endothelial cells is mediated..., Ma [/bib_ref] , is highly enriched in lipid raft fractions, wherein it may interact with β2-GPI. It is important to note that, although cell-surface ANXA2 lacks an intracellular tail and is unable to induce intracellular signal transduction by itself, it can act as a binding partner for intracellular surface molecules in lipid rafts [bib_ref] Cooperative binding of annexin A2 to cholesterol-and phosphatidylinositol-4,5-bisphosphate-containing bilayers, Drücker [/bib_ref] [bib_ref] Lipid raft endocytosis and exosomal transport facilitate extracellular trafficking of annexin A2, Valapala [/bib_ref]. The involvement of TLR2, as well as TLR4 and the interaction between β2-GPI and ANXA2 within lipid rafts supports the view that TLRs act as "adaptor" proteins with consequent IRAK phosphorylation and activation of NF-κB. These findings indicated that lipid rafts play a key role in the signal transduction pathway induced by anti-β2-GPI antibodies, and that raft-dependent anti-β2-GPI antibody triggering resulted in the release of either TNF-α and TF, which may contribute to the pathogenesis of thrombosis in APS [bib_ref] Thrombosis in primary antiphospholipid syndrome: A pivotal role for monocyte tissue factor..., Cuadrado [/bib_ref] [bib_ref] Beta-2-glycoprotein I expression on monocytes is increased in anti-phospholipid antibody syndrome and..., Conti [/bib_ref].
Over the years, a molecular mimicry has been shown between β2-GPI and bacterial antigens, or other various infectious agents [bib_ref] A peptide that shares similarity with bacterial antigens reverses thrombogenic properties of..., Pierangeli [/bib_ref]. Interestingly, an epitope of P gingivalis RNA polymerase s-70 factor shares a complete identity with a peptide of β2-GPI domain I.
Recent evidence showed the presence of autoantibodies specific to this peptide of β2-GPI, which shares a high homology with an extracellular epitope of TLR4, in APS sera. Indeed, TLRs are important components of innate immunity, recognizing specific microbial products and driving the inflammatory response [bib_ref] Toll-like receptors: Key mediators of microbe detection, Underhill [/bib_ref]. Antibodies directed to the epitope shared by β2-GPI and TLR4 induced IRAK phosphorylation and NF-kB translocation through the triggering of the TLR4 signaling cascade within lipid rafts, thereby promoting both VCAM expression on endothelial cells and TNF-a release from monocytes responsible for a proinflammatory phenotype [bib_ref] Autoantibodies specific to a peptide of β2-glycoprotein I cross-react with TLR4, inducing..., Colasanti [/bib_ref].
Recent evidence showed that anti-β2-GPI antibodies may also trigger a similar signal transduction pathway in human platelets, which involves IRAK phosphorylation and NF-κB activation followed by TF expression, indicating that platelets may also play a role in the pathogenetic mechanism of APS through lipid rafts [bib_ref] Tissue factor over-expression in platelets of patients with anti-phospholipid syndrome: Induction role..., Capozzi [/bib_ref].
Indeed, TF expression may be induced through the simultaneous activation of NF-κB proteins (via the p38 MAPK pathway) and of the MEK-1/ERK pathway [bib_ref] Antiphospholipid antibodies from patients with the antiphospholipid syndrome induce monocyte tissue factor..., Lopez-Pedrera [/bib_ref]. This may reflect the presence of more than one synergic activating pathway.
## "new" signaling in the immunopathogenesis of aps
Although numerous aspects of the molecular mechanisms involved in the immunopathogenesis of APS remain still unknown, recent papers have provided new knowledge of the underlying additional signaling pathways triggered by aPLs, and often these pathways connect the innate immune system and the coagulation cascade [bib_ref] Inhibition of the mTORC pathway in the antiphospholipid syndrome, Canaud [/bib_ref] [bib_ref] Lipid presentation by the protein C receptor links coagulation with autoimmunity, Müller-Calleja [/bib_ref]. Following some observations that protease activated receptor 2 (PAR-2) inhibition prevents the expression of TF induced by aPLs [bib_ref] Tissue factor pathway inhibitor primes monocytes for antiphospholipid antibody-induced thrombosis, Müller-Calleja [/bib_ref] [bib_ref] Neutrophil activation by the tissue factor/Factor VIIa/PAR2 axis mediates fetal death in..., Redecha [/bib_ref] and that LRP6 has been identified as a novel co-receptor of PAR-2 [bib_ref] Bar-Shavit, R. Low-density lipoprotein receptor-related protein 6 is a novel coreceptor of..., Nag [/bib_ref] , we described an additional signaling in which anti-β2-GPI antibodies trigger the LRP6 pathway, with consequent phosphorylation of β-catenin. LRP6 belongs to the same family as LRP8, a putative receptor of anti-β2-GPI antibodies, but also appears to bind oxide phospholipids involved in pathological conditions such as atherosclerosis and inflammation [bib_ref] Oxidized phospholipids are ligands for LRP6, Wang [/bib_ref]. LRP6, a type I transmembrane receptor of LDL receptor-related proteins, is a pivotal co-receptor in numerous biological processes. LRP6 binds different Wnt ligands [bib_ref] LRP6 mediated signal transduction pathway triggered by tissue plasminogen activator acts through..., Riitano [/bib_ref] , but also several Wnt antagonists, such as Dickkopf 1 (DKK1) [bib_ref] Novel mechanism of Wnt signalling inhibition mediated by Dickkopf-1 interaction with LRP6/Arrow, Bafico [/bib_ref] [bib_ref] LDL-receptor-related protein 6 is a receptor for Dickkopf proteins, Mao [/bib_ref] [bib_ref] SOST is a ligand for LRP5/LRP6 and a Wnt signaling inhibitor, Semënov [/bib_ref]. β-catenin, which represents the key effector of this signaling pathway, is constantly degraded; however, in the presence of Wnt, it translocases to the nucleus, inducing the expression of an array of genes downstream [bib_ref] Bar-Shavit, R. Low-density lipoprotein receptor-related protein 6 is a novel coreceptor of..., Nag [/bib_ref] [bib_ref] Faussner, A. β-Catenin-dependent pathway activation by both promiscuous "canonical" WNT3a-, and specific..., Ring [/bib_ref]. Haack et al. [bib_ref] Receptor/Raft Ratio Is a Determinant for LRP6 Phosphorylation and WNT/β-Catenin Signaling, Haack [/bib_ref] suggested lipid raft involvement in the Wnt/β-catenin pathway. Our findings indicated that anti-β2-GPI antibodies interact with their receptor complex within lipid rafts, leading to the formation of the complex β2-GPI-LRP6-PAR-2. This hypothesis was strongly supported by coimmunoprecipitation experiments, which revealed that β2-GPI coupled with LRP6 after anti-β2-GPI antibodies treatment. Moreover, the anti-β2-GPI-induced TF expression was prevented by treatment with MβCD, a raft-affecting drug, as well as by treatment with DKK1, a selective inhibitor of LRP6. Thus, this additional signaling pathway seems to be involved in the induction of procoagulant phenotype of endothelial cells.
Thrombosis is the major disease mechanism of APS and vasculopathy, enhanced mainly by severe intimal hyperplasia; it can also play a role in vascular occlusions and pregnancy morbidity [bib_ref] Antiphospholipid arterial vasculopathy, Alarcón-Segovia [/bib_ref] [bib_ref] Vasculopathy and arterial stenotic lesions in the antiphospholipid syndrome, Christodoulou [/bib_ref]. Recently, another signaling related to vasculopathy of APS has been described, in which aPLs directly activate the mammalian target of the rapamycin complex (mTORC) pathway in intrarenal vessels of patients with primary and secondary APS nephropathy [bib_ref] Inhibition of the mTORC pathway in the antiphospholipid syndrome, Canaud [/bib_ref]. Moreover, patients with APS nephropathy who required transplantation and were receiving Sirolimus, an inhibitor of mTORC [bib_ref] Inhibition of intimal thickening after balloon angioplasty in porcine coronary arteries by..., Gallo [/bib_ref] , displayed decreased vascular proliferation. These data have been confirmed by in vitro experiments, where the intensity of mTORC activation in endothelial cells has been shown to be correlated with aPL titers. Furthermore, the exposure of endothelial cells to Sirolimus for 1 h completely inhibited the phosphorylation of S6 ribosomal protein (S6RP) induced by aPLs, but not the phosphorylation of AKT (Ser473). However, cells treated with aPLs and exposed to Sirolimus for 48 h showed reduced phosphorylation of both S6RP and AKT (Ser473). These results are consistent with the activation of both mTORC1 and mTORC2 in endothelial cells, leading to their proliferation and the proliferation of the surrounding vascular smooth muscle cells (VSMC). Moreover, to explain both thrombotic and pregnancy complications of APS, a complex signaling pathway triggered by the interaction between aPLs and the endothelial protein C receptor (EPCR), which is bound to lyso-bis-phosphatidic acid (LBPA), has been recently described [bib_ref] Lipid presentation by the protein C receptor links coagulation with autoimmunity, Müller-Calleja [/bib_ref]. Under normal physiological conditions, EPCR has an important anti-thrombotic and anti-inflammatory effect. EPCR belongs to the MHC class I/CD1 family, and shares with these molecules a feature of lipid exchanging and loading through endosomal recycling [bib_ref] Lysosomal glycosphingolipid recognition by NKT cells, Zhou [/bib_ref] [bib_ref] Exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections, Mattner [/bib_ref]. After the lipid-reactive aPL binding, EPCR internalizes itself. This effect implies the exchange of EPCR-bound phosphatidylcholine, a membrane phospholipid, with LBPA, a late endosomal lipid. aPLs, by binding the EPCR-LBPA complex, may induce the coagulation cascade and activate monocytes and endothelial cells into a pro-coagulant phenotype. The interaction of aPLs with EPCR-LBPA triggers the activation of embryonic trophoblast cells. Moreover, the EPCR-LBPA complex promotes the production of α-interferon by dendritic cells, which eventually sustains a TLR7-dependent expansion of B1a cells and enhances production of aPLs. B1 cells are a rare B lymphocyte subpopulation with unique cell surface antigens that spontaneously secrete IgM. The expansion of B1a cells may be responsible for producing lipid-reactive aPLs in a vicious cycle that enhances the damage mechanisms in APS. The subsequent upregulation of TNF-α and TF plays a key role in the pathogenesis of thrombosis and pregnancy complications. In vivo, blocking EPCR-LBPA signaling in Tlr7−/− mice prevents the development of lipid reactive aPLs and remarkably attenuates thrombosis, fetal loss, and kidney damage [bib_ref] Lipid presentation by the protein C receptor links coagulation with autoimmunity, Müller-Calleja [/bib_ref].
However, several events are likely to play a role in pathogenesis of APS. Recent studies have focused on new signaling pathways responsible for complications of APS that could be useful in understanding the pathogenesis of APS, but also in improving the management of APS patients.
## Potential therapeutic strategies through lipid rafts
The well-known multifactorial nature of the pathogenic mechanisms of APS opens a broad panorama of potential target molecules. However, at present, therapeutic strategies are aimed at the symptom, and therefore essentially at the treatment of thromboinflammation. In fact, today, the long-term management of patients with APS essentially involves alternative vitamin K antagonists, such as the new oral anticoagulant agents, direct and indirect thrombin inhibitors, hydroxychloroquine, and TF inhibitors [bib_ref] Use of new oral anticoagulants in antiphospholipid syndrome, Arachchillage [/bib_ref] [bib_ref] update: Hopkins lupus cohort, Fangtham [/bib_ref] [bib_ref] Characterization of monocyte tissue factor activity induced by IgG antiphospholipid antibodies and..., Zhou [/bib_ref] [bib_ref] Tissue factor in cardiovascular disease pathophysiology and pharmacological intervention, Holy [/bib_ref].
The search for new therapies that can guarantee clinical efficacy and at the same time avoid too heavy an impact on the patient's quality of life is very open. For reasons of efficacy and safety, also linked to the impact of long-term therapy on the patient's daily life, adherence to treatment and possible food interactions, it is necessary to experiment with new therapeutic strategies. The most recent discoveries on the pathogenetic mechanisms of APS show us new avenues for targeted therapies.
In this concern, as reported in [fig_ref] Figure 1: Figure 1 [/fig_ref] , various signaling pathways involved in the procoagulant effect of aPLs act through lipid rafts. Moreover, the signal transduction pathway triggered by anti-β2-GPI antibodies can be strongly influenced by the structure and function of the lipid raft; in fact, modifications in the size and/or structure of lipid rafts would lead to impaired stability and function of these structures and, consequently, could cause signaling abnormalities [bib_ref] Lipid rafts as a therapeutic target, Sviridov [/bib_ref].
Thus, targeting lipid rafts is a rapidly growing therapeutic approach for the treatment of various diseases, from neurodegeneration and neuropathic pain to cancer, infections and atherosclerosis. The approach's aims to affect lipid rafts can be different; they may be through raising the levels of natural cholesterol acceptors (i.e., HDL) or their mimetic counterparts, or facilitating the efflux of cholesterol with suitable transporters [bib_ref] Phosphorylation and stabilization of ATP binding cassette transporter A1 by synthetic amphiphilic..., Arakawa [/bib_ref] [bib_ref] Apolipoprotein AI and high-density lipoprotein have anti-inflammatory effects on adipocytes via cholesterol..., Umemoto [/bib_ref] [bib_ref] Apolipoprotein A-I attenuates palmitate-mediated NF-kappaB activation by reducing toll-like receptor-4 recruitment into..., Cheng [/bib_ref] [bib_ref] Neutrophil activation is attenuated by high-density lipoprotein and apolipoprotein A-I in in..., Murphy [/bib_ref] [bib_ref] Cyclodextrin promotes atherosclerosis regression via macrophage reprogramming, Zimmer [/bib_ref]. However, the most direct approach consists of depleting the structural lipids responsible for the stability of the raft, such as glycosphingolipids cholesterol or sphingomyelin [bib_ref] Lipid rafts as a therapeutic target, Sviridov [/bib_ref]. In this concern, inhibition of glycosphingolipid biosynthesis in vitro with a clinically approved inhibitor (N-butyldeoxynojirimycin) corrected CD4 + T cell signaling and functional defects, and decreased autoantibody production in SLE patients [bib_ref] Normalizing glycosphingolipids restores function in CD4+ T cells from lupus patients, Mcdonald [/bib_ref].
However, the most common handling approach for affecting lipid rafts includes molecules that are capable of physically sequestering and subtracting cholesterol from the cell membrane [bib_ref] Intrathecal 2-hydroxypropyl-beta-cyclodextrin decreases neurological disease progression in Niemann-Pick disease, type C1: A..., Ory [/bib_ref] [bib_ref] Tumor targeting and lipid rafts disrupting hyaluronic acid-cyclodextrin-based nanoassembled structure for cancer..., Lee [/bib_ref]. In particular, cyclodextrins have been used for some time, both as a food additive (therefore safe for consumption) and as an additive to various pharmacological preparations to increase the stability, solubility and bioavailability of many drugs with anti-inflammatory properties [bib_ref] Normalizing glycosphingolipids restores function in CD4+ T cells from lupus patients, Mcdonald [/bib_ref]. Cholesterol and sphingolipids can also be depleted from the cell, using inhibitors of their biosynthesis such as statins and Miglustat, respectively. Statins, competitive inhibitors of HMG-CoA reductase, a ratelimiting enzyme of the cholesterol biosynthesis pathway, have long been widely used for the treatment of hypercholesterolemia and represent a concrete possibility of modulating the activity of lipid rafts. Statins are the ideal example of connection with the classical strategy based on membrane cholesterol depletion. They are a well-known class of cholesterol lowering agents and possess several pleiotropic effects (i.e., cholesterol-independent), including the ability to influence the organization of artificial and biological membranes.
Indeed, they are cholesterol-regulating agents that exert anti-inflammatory, immunoregulatory, and anti-thrombotic properties [bib_ref] Inhibition of the thrombogenic and inflammatory properties of antiphospholipid antibodies by fluvastatin..., Ferrara [/bib_ref] [bib_ref] Statins prevent endothelial cell activation induced by antiphospholipid (anti-beta2-glycoprotein I) antibodies: Effect..., Meroni [/bib_ref] [bib_ref] Reduction in early stroke risk in carotid stenosis with transient ischemic attack..., Merwick [/bib_ref]. In vitro and in vivo studies (animal models with APS) demonstrated that statins can inhibit the activation of endothelial cells by aPLs and prevent the overexpression of TF, IL-6 and adhesion molecules [bib_ref] A randomized trial of rosuvastatin in the prevention of venous thromboembolism, Glyn [/bib_ref] [bib_ref] Fluvastatin inhibits up-regulation of tissue factor expression by antiphospholipid antibodies on endothelial..., Ferrara [/bib_ref] [bib_ref] Pravastatin prevents miscarriages in antiphospholipid antibody-treated mice, Girardi [/bib_ref] [bib_ref] Global effects of fluvastatin on the prothrombotic status of patients with antiphospholipid..., Lopez-Pedrera [/bib_ref]. Thus, statins (by inhibition of cholesterol synthesis) or cyclodextrins (by depletion of membrane cholesterol), through affecting cholesterol levels and disrupting lipid rafts, could effectively inhibit the aPLs signaling pathway's activation [fig_ref] Figure 2: Figure 2 [/fig_ref].
In conclusion, the progress made in recent years in understanding the biological significance of lipid rafts is indisputable. However, further investigations needed to be conducted, including on the interplay between lipids and membrane proteins in affect membrane organization, in order to provide new therapeutic strategies.
However, targeting the pathogenic pathways of APS development would be crucial, and various studies are underway. A deeper understanding of the immunological mechanisms at the basis of APS is still needed in order to improve the classical anti-thrombotic approaches with "new" immunomodulatory drugs.
# Conclusions
Emerging data pointed out the key role of lipid rafts, specific microdomains on the plasma membrane of endothelial cells, monocytes and platelets, in the signal transduction pathway(s) implied in the pathogenesis of APS. Knowledge of these pathways and clari-fication of the role of lipid rafts might present new perspectives on the deepening of the immunopathogenesis of the syndrome and on the identification of "new" pharmacological targets that will allow us to move towards personalized medicine.
## Conflicts of interest:
The authors declare no conflict of interest.
[fig] Figure 1: Figure 1. Pathogenetic mechanisms contributing to thrombosis in APS and possible molecular targets. Molecular mechanisms of aPLs, in particular of anti-β2-GPI antibodies, involve the cooperation of endothelial cells, monocytes and platelets with a production of adhesion molecules, proinflammatory cytokines and tissue factor, leading to a proinflammatory and procoagulant state. Anti-β2-GPI antibodies may also activate neutrophils by inducing the formation of NETs important to propagate inflammation and contribute to thrombosis. Inflammation and complement activation play a central role in upregulating cell activation in the pathophysiology of APS. These pathogenetic pathways underlying clinical manifestations of APS can be exploited as targets of immunomodulatory therapeutic strategies to be used in APS patients. Beta2-glycoprotein I (β2-GPI); Annexin 2 (ANXA2); toll-like receptors (TLRs); C3a receptor (C3a-R); C5a receptor (C5a-R); cholesterol (Chol); endothelial protein C receptor (EPCR); lyso-bis-phosphatidic acid (LBPA); myeloid differentiation factor 88 (MyD88); mammalian target of rapamycin complex (mTOR); nuclear factor-kappa B (NF-κB); neutrophil extracellular traps (NETs); vascular cellular adhesion molecule (VCAM-1); intercellular adhesion molecule-1 (ICAM-1); tumor necrosis factor alpha (TNFα); tissue factor (TF); interleukin (IL). [/fig]
[fig] Figure 2: Figure 2. Signaling pathways induced by anti-β2-GPI antibodies. Scheme depicting the major intracellular signaling pathways triggered by anti-β2-GPI antibodies via TLRs and/or LRP6 through lipid rafts (dotted circle), implicated in the procoagulant endothelial cell phenotype characterized by the expression of TF. Lipid raft-targeted drugs act on cholesterol either as cholesterol depletors (MβCDs) or cholesterol synthesis inhibitors (statins). Annexin A2 (ANXA2); anti-β2-glycoprotein I (anti-β2-GPI antibodies); beta-catenin (β-catenin); cholesterol (Chol); interleukin-1 receptor-associated kinases (IRAKs); LDL receptor related protein 6 (LRP6); methyl-β-cyclodextrin (MβCD); myeloid differentiation factor 88 (MyD88); nuclear factor-kappa B (NF-κB); protease activated receptor-2 (PAR-2); tissue factor (TF); toll-like receptors (TLRs); TNF receptor associated factor 6 (TRAF6). [/fig]
[fig] Author: Contributions: R.M., M.S. and A.C. contributed to the conception and design of the work. R.M., M.S., A.C., V.M. and G.R. drafted the work. T.G., D.C. and A.L. revised it critically for important intellectual content. A.C. and V.M. contributed to the acquisition of published literature and to the original iconography. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data sharing not applicable. [/fig]
|
Antibacterial mechanisms identified through structural systems pharmacology
Background: The growing discipline of structural systems pharmacology is applied prospectively in this study to predict pharmacological outcomes of antibacterial compounds in Escherichia coli K12. This work builds upon previously established methods for structural prediction of ligand binding pockets on protein molecules and utilizes and expands upon the previously developed genome scale model of metabolism integrated with protein structures (GEM-PRO) for E. coli, structurally accounting for protein complexes. Carefully selected case studies are demonstrated to display the potential for this structural systems pharmacology framework in discovery and development of antibacterial compounds. Results: The prediction framework for antibacterial activity of compounds was validated for a control set of well-studied compounds, recapitulating experimentally-determined protein binding interactions and deleterious growth phenotypes resulting from these interactions. The antibacterial activity of fosfomycin, sulfathiazole, and trimethoprim were accurately predicted, and as a negative control glucose was found to have no predicted antibacterial activity. Previously uncharacterized mechanisms of action were predicted for compounds with known antibacterial properties, including (1-hydroxyheptane-1,1-diyl)bis(phosphonic acid) and cholesteryl oleate. Five candidate inhibitors were predicted for a desirable target protein without any known inhibitors, tryptophan synthase β subunit (TrpB). In addition to the predictions presented, this effort also included significant expansion of the previously developed GEM-PRO to account for physiological assemblies of protein complex structures with activities included in the E. coli K12 metabolic network.Conclusions:The structural systems pharmacology framework presented in this study was shown to be effective in the prediction of molecular mechanisms of antibacterial compounds. The study provides a promising proof of principle for such an approach to antibacterial development and raises specific molecular and systemic hypotheses about antibacterials that are amenable to experimental testing. This framework, and perhaps also the specific predictions of antibacterials, is extensible to developing antibacterial treatments for pathogenic E. coli and other bacterial pathogens.
# Background
Structural systems pharmacology [bib_ref] Structural systems pharmacology: a new frontier in discovering novel drug targets, Tan [/bib_ref] is the study of drug action through characterization of proteome-wide drugtarget interactions and their systemic consequences. A previously developed local structure homology-based approach to predicting ligand binding pockets (SMAP) [bib_ref] Detecting evolutionary relationships across existing fold space, using sequence order-independent profile-profile alignments, Xie [/bib_ref] [bib_ref] Drug discovery using chemical systems biology: identification of the protein-ligand binding network..., Xie [/bib_ref] [bib_ref] SMAP-WS: a parallel web service for structural proteome-wide ligand-binding site comparison, Ren [/bib_ref] has been applied efficaciously in multiple contexts to study pharmacological phenomena [bib_ref] Drug discovery using chemical systems biology: repositioning the safe medicine Comtan to..., Kinnings [/bib_ref] [bib_ref] Drug off-target effects predicted using structural analysis in the context of a..., Chang [/bib_ref] [bib_ref] The Mycobacterium tuberculosis drugome and its polypharmacological implications, Kinnings [/bib_ref] [bib_ref] Raloxifene attenuates Pseudomonas aeruginosa pyocyanin production and virulence, Ho Sui [/bib_ref]. The recent development of a structural biology resource with which to study physiological stresses upon the proteome of Escherichia coli K12 MG1655 metabolism [bib_ref] Structural systems biology evaluation of metabolic thermotolerance in Escherichia coli, Chang [/bib_ref] has enabled a diversity of potential applications. Thus, we applied the SMAP methodology and the E. coli metabolic genome-scale model integrated with protein structures (GEM-PRO), to analyze and predict antibacterial effects of chemical compounds. E. coli K12, although not pathogenic under normal circumstances, is a well-characterized laboratory model for enteropathogenic bacteria that infect humans. Thus methods, and perhaps even some specific predictions of antibacterial properties made in this study, are extensible to pathogenic E. coli and other bacterial pathogens. In addition to the integrative framework presented in this study for structural systems pharmacology, this effort also included significant expansion of the previously developed GEM-PRO to account for physiological assemblies of protein complex structures with activities accounted for in the E. coli K12 metabolic network iJO1366 [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref]. Results from this study show promising proof of principle for such an analysis framework and raise specific molecular and systemic hypothesis about antibacterials that are amenable to experimental testing.
# Results
## Expansion of gem-pro to include protein complexes
Many proteins do not act as monomers in the cell but as part of multimeric protein complexes that may include proteins encoded by one or several distinct genes. The previously constructed Escherichia coli genomescale model integrated with protein structures (GEM-PRO) [bib_ref] Structural systems biology evaluation of metabolic thermotolerance in Escherichia coli, Chang [/bib_ref] considered proteins solely as single-peptide chains. As a result, we sought to expand the scope of GEM-PRO to account for the structure of protein complexes. The structures of protein complexes are complementary to the existing single-peptide chain structures already included in the E. coli GEM-PRO. The objective was to best represent the physiological assemblies of metabolic enzyme complexes, that is, the best structural representation of the active form of enzyme complexes in vivo. A conceptual representation of this expansion with respect to the example reaction of glucosamine-1phosphate N-acetyltransferase (G1PACT) is displayed in [fig_ref] Figure 1: phan synthase β subunit [/fig_ref] ; in this case, the physiologically active form of the GlmU enzyme is a homotrimer.
There are 1106 functional enzymatic complexes [bib_ref] EcoCyc: a comprehensive database of Escherichia coli biology, Keseler [/bib_ref] known to form among the proteins accounted for in iJO1366 [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref]. The overall coverage of complexes in this GEM-PRO is 519 out of the 1106 known complexes [fig_ref] Figure 1: phan synthase β subunit [/fig_ref] ; Of these 519 complexes, 426 are completely represented with accurate subunit stoichiometry by a single structure in the expanded GEM-PRO, and another 93 complexes are partially represented by structures, which may not include all distinct polypeptide subunits of the complex or may have incomplete subunit stoichiometry. This effort yielded 527 individual protein structure files, 149 of which were redundant with structures contained in the previously developed GEM-PRO [bib_ref] Structural systems biology evaluation of metabolic thermotolerance in Escherichia coli, Chang [/bib_ref]. As is clear from [fig_ref] Figure 1: phan synthase β subunit [/fig_ref] , a slight majority of known complexes are not represented at all in the complex expansion to the GEM-PRO. A combination of the EcoCyc database [bib_ref] EcoCyc: a comprehensive database of Escherichia coli biology, Keseler [/bib_ref] , PDB structure curation [bib_ref] The protein data bank, Berman [/bib_ref] , computational assessment of symmetry operations on the asymmetric unit of protein crystals [bib_ref] Inference of macromolecular assemblies from crystalline state, Krissinel [/bib_ref] , and literature review were used to identify a consensus for the most physiologically accurate assemblies currently possible (see Additional file 1: [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref]. These assemblies were distributed among different classes of oligomeric states: monomers, homomultimers, and heteromultimers [fig_ref] Figure 1: phan synthase β subunit [/fig_ref].
The monomers directly overlap with contents previously reconstructed [bib_ref] Structural systems biology evaluation of metabolic thermotolerance in Escherichia coli, Chang [/bib_ref].
## Structure-based prediction of protein targets of antibacterials
The expanded E. coli GEM-PRO was employed prospectively to explore possible currently unknown antibacterial properties. Two pipelines were established to screen for different types of antibacterial associations . Protein targets for antibacterials with unknown mechanisms of action, compounds known to have antibacterial effects but without known molecular targets, were predicted , and anti-metabolite compounds were also predicted as novel antibacterials to target orphan protein targets without known inhibitors . Protein-ligand targeting was predicted using the previously developed SMAP method [bib_ref] SMAP-WS: a parallel web service for structural proteome-wide ligand-binding site comparison, Ren [/bib_ref]. Some negative and positive control antibacterial compounds were also screened, for which there is existing data on antibacterial properties and established mechanisms of action within metabolism.
A subset of the results of these screens are summarized in [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref] , including novel predicted compound targets and those that displayed antibacterial properties through simulation of inhibition in the metabolic model (described later); the full set of SMAP predictions is presented in Additional file 2: [fig_ref] Table 2: Metabolic model performance in predicting antibacterial effects [/fig_ref].
In the negative control screen for glucose (BGC) SMAP predicted that glucose significantly binds to 7 individual metabolic E. coli proteins and 2 protein complexes, one of which is a known target (MglB). Using less stringent significance criteria for the SMAP p-value revealed a second known target (Glk). Some of these targets are expected because glucose is a known substrate of these proteins. Although SMAP does not predict significant binding of glucose to glycogen phosphorylase (GlgP), for which it is a known inhibitor, this protein does rank 4 th of 3234 structures for one screen (p-value = 9.55 × 10 -3 ). Because we assume that glucose binding targets are the most extensively characterized of all compounds included in this study, these negative control screens were also used to examine the false positive rate of SMAP predictions of ligand binding. Using stated significance criteria (see methods), 9 false positive and 3207 true negative predictions resulted, corresponding to a false positive rate of 0.0028.
Of the positive antibacterial controls, the top SMAP hit for the sulfonamide 4-amino-N-(1,3-thiazol-2-yl) benzenesulfonamide (YTZ) is the known primary target, dihydropteroate synthase (FolP). Two other positive controls, fosfomycin (FCN) and trimethoprim (TOP), were predicted by SMAP to bind significantly to a number of proteins [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref] , none of which were known targets, leaving these predictions as unresolved but nevertheless putative targets defining unknown mechanisms leading to an antibacterial effect, described further below. The positive control 2,2′-methanediylbis(3,4,6-trichlorophenol) (H3P) was not predicted by SMAP to significantly bind any proteins; although the known primary target (FabI) was ranked 122 nd out of 3233 protein structures. The experimentallycharacterized binding mode of H3P co-crystalized with bovine glutamate dehydrogenase (GDH) is as a ring consisting of six H3P molecules [bib_ref] Novel inhibitors complexed with glutamate dehydrogenase: allosteric regulation by control of protein..., Li [/bib_ref] , each molecule interacting both with the GDH homohexamer and with two other neighboring H3P molecules. This complex binding mode may explain the lower than expected significance of SMAP hits for known H3P targets, as the template for the binding site used for the SMAP screen did not capture the six-molecule ring binding mode.
The antibacterial 4-(aminomethyl)benzoic acid (4AZ), with unknown action mechanism, was not predicted to significantly bind to any metabolic proteins. Intriguingly, the two other antibacterials with unknown mechanisms of action screened in this study, (1-hydroxyheptane-1,1-diyl) bis(phosphonic acid) (028) and cholesteryl oleate (2OB), were both predicted as significant binders by SMAP to multiple metabolic proteins [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref] , suggesting possible mechanisms for their antibacterial activity.
Of the three screens aiming to identify anti-metabolite inhibitors of known essential genes in E. coli, SMAP predicted 5 candidate inhibitors for the trypto- [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref]. SMAP screens for inhibitors of erythronate-4-phosphate dehydrogenase (PdxB) and orotate phosphoribosyltransferase (PyrE) failed to predict any significant candidate inhibitors. Several other known metabolic targets of the control compounds were not predicted by SMAP. In our preliminary control screens, it was hypothesized that there may exist distinct binding pocket motifs for an individual compound such that using a single protein template to search for other targets may not identify all true targets of a compound. Expanding the number of search templates for a single compound, as was done for BGC, FCN, and TOP, indeed identified more significant known targets, supporting this hypothesis.
To assess the relative accuracy of SMAP in predicting true positive protein-ligand interactions, we performed statistical analysis of the entire set of SMAP results, including insignificant calls. Mann Whitney U-tests were run on the ranked lists of SMAP predictions with respect to each template protein structure, yielding inconsistently statistically significant p-values for some compounds [fig_ref] Figure 3: SMAP performance in recalling true positives [/fig_ref]. This result too supports that different binding motifs may exist for an individual compound, as is most apparent for BGC and TOP, which show the widest range of p-values. To highlight the overall efficacy of SMAP in predicting true positives, the results from all screens for a particular compound were combined by considering only the top rank number for each protein structure, whether a known target or not. It is apparent from [fig_ref] Figure 3: SMAP performance in recalling true positives [/fig_ref] that the examples BGC, FCN, TOP, and H3P all noticeably support SMAP's predictive accuracy; however, the stringency of significance criteria used may obscure this ability for many protein-ligand interactions. Because there is no obvious a priori approach to choosing a single structural template for screening a compound that may bind to multiple distinct motifs, our results suggest that using as wide an array of diverse templates as appropriate should be considered when running SMAP screens. This phenomenon may explain some of the false negative SMAP predictions for controls in this study. Antibacterial prediction pipelines. (A) Screening causal targets for antibacterial activity of input compounds. Seeded with at least one structure of the compound of interest bound to a known target and the GEM-PRO to represent the functional proteome, SMAP is run to predict binding partners within the GEM-PRO. The potential for these predicted binding events to inhibit protein activity is then evaluated based on binding site overlap with native functional sites annotated in the GEM-PRO. Targets exhibiting overlap of antibacterial binding sites and functional sites are then evaluated for their inhibition growth phenotype in the GEM-PRO using the COBRA Toolbox. The inhibitable protein targets leading to deleterious growth phenotypes comprise predictions of causal targets for antibacterial activity. (B) Screening inhibitors of desired antibacterial target protein(s). Seeded with the GEM-PRO, metabolic simulations may be performed using the COBRA Toolbox to predict phenotypic impacts of protein inhibition to identify potential antibacterial target protein(s); alternatively, desirable targets may be chosen based on experimental results, such as gene-knockout phenotypes. To search for inhibitors of the chosen targets, the native functional sites of the proteins are identified, as in the GEM-PRO, and passed to SMAP to screen ligand-binding pockets of structures included in the PDB, searching for significant local structural matches. Significant matches comprise potential inhibitors of the chosen target proteins, expected to hold antibacterial properties. implications with respect to two protein complexes, not exhibited with respect to the complex subunits in isolation. The predicted 2OB binding site on the cytochrome bo terminal oxidase appears at the interaction site between CyoB and CyoC. The 2OB binding site also overlapped with the heme binding sites of the SdhC and SdhD subunits of the succinate dehydrogenase complex as well as the protein-protein interaction region between these subunits. These last few predictions speak to the importance of the complex expansion of the GEM-PRO, without which such molecular predictions involving multiple subunit interfaces would not have been possible.
## A b
## Simulation of phenotypes from antibacterial target inhibition
Finally, we turned to the metabolic network portion of the E. coli GEM-PRO, iJO1366 [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref] , to simulate the outcomes of known and predicted binding events leading to inhibition of protein activity and determine whether or not these events may be detrimental to growth. First, we tested the ability of the model to accurately predict the phenotypic impact caused by inhibition of known targets of all control compounds [fig_ref] Table 2: Metabolic model performance in predicting antibacterial effects [/fig_ref]. Inhibition of all known and predicted binding targets of BGC led to no decrease in growth phenotype, accurately predicting the known outcome of the negative control. Inhibition of positive control targets led to no growth or reduced growth rates in the model. In combination, the collective inhibition of all known targets for each positive control compound led to complete growth inhibition, but remarkably, most of these targets individually also led to complete loss of growth if inhibited, only failing to predict deleterious growth phenotypes upon inhibition of FbaA, TolC, and FolA individually. The effects of inhibition of SMAP-predicted targets were then evaluated in the model. Each of the individual predicted protein targets reported in [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref] exhibited decreased or no growth upon full inhibition in simulation. These predictions helped to pare down the list of significant SMAP predictions to those that satisfy both lines of evidence for antibacterial effects. With the exception of the FolP-YTZ binding interaction, all of the interactions reported in [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref] are previously unknown, which suggests that in the case of positive control compounds, we may have uncovered previously unknown antibacterial targets. For the antibacterial compounds with unknown mechanisms of action, we predicted that inhibition of IspA and IspB by 028 leads to decreased growth rate and that inhibition of 14 individual proteins and 2 protein complexes by 2OB leads to decreased growth rate. Further details of the specific pathways impacted by these inhibitory activities were investigated in the flux balance model. The mechanistic models of antibacterial activity of 028, 2OB, and potential inhibitors of TrpB are summarized in [fig_ref] Figure 4: Predicted antibacterial mechanisms [/fig_ref] , with more detailed network flux maps provided in Additional file 3: [fig_ref] Figure 1: phan synthase β subunit [/fig_ref]. In the mechanistic model for 028 [fig_ref] Figure 4: Predicted antibacterial mechanisms [/fig_ref] , IspA and IspB are inhibited leading to decreased isoprenoid synthesis activity and ultimately no model growth. The mechanistic model for 2OB [fig_ref] Figure 4: Predicted antibacterial mechanisms [/fig_ref] includes inhibition of several proteins (PheA, AcpP, EntA, and AtpB) and protein complexes (cytochrome bo terminal oxidase and succinate dehydrogenase) participating in a variety of metabolic pathways (amino acid synthesis, lipid synthesis, enterochelin metabolism, and oxidative phosphorylation) ultimately leading to no model growth.
We also tested if inhibition of the individual protein targets predicted by gene-knockout phenotypes to be effective antibacterial targets leads to growth deficits in the model and found that all three individual inhibitions lead to no growth in the model [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref]. However, as previously mentioned, our SMAP screens only predicted potential inhibitors of TrpB. The mechanistic model for antibacterial activity of these compounds is presented in [fig_ref] Figure 4: Predicted antibacterial mechanisms [/fig_ref] , where any of F6F, PLT, 7MN, IDM, or PLS are expected to inhibit TrpB activity, thereby inhibiting tryptophan synthesis and leading to no growth in simulation.
# Discussion
In this study, we have demonstrated the first structural systems pharmacology antibacterial screens for the model bacterium E. coli. This effort was enabled in part through the expansion of the E. coli GEM-PRO to include protein complexes. In this attempt at reconstruction of metabolic protein complexes, we chose to utilize only those structures supported by strong experimental evidence; however, the XylA YlaD scope of this reconstruction could be further expanded through modeling of protein complex structures, as has been attempted by others recently [bib_ref] Structure-based prediction of protein-protein interactions on a genome-wide scale, Zhang [/bib_ref]. Our previous and current efforts at reconstructing the E. coli metabolic GEM-PRO have enabled in silico exploration of diverse forms of physicochemical stress, but much broader expansions are likely to emerge and enable still more diverse avenues of investigation. One important lesson learned from this study is that availability of only a few static structures to represent proteins may limit the sensitivity of ligand binding prediction. Prospectively, molecular dynamics simulations could be used to generate ensembles of structures [bib_ref] Improving structure-based function prediction using molecular dynamics, Glazer [/bib_ref] for each protein to perhaps include the conformations necessary to uncover more binding interactions, lending greater sensitivity to the prediction approach. Generating such ensembles for the proteins included in this study would be a substantial effort given the high number of protein structures included in this GEM-PRO and the long simulation time scales necessary to model the large conformational changes often important for ligand binding [bib_ref] Bioinformatics and variability in drug response: a protein structural perspective, Lahti [/bib_ref]. The expected resultant increase in query database size would also dramatically increase SMAP runtime. Nevertheless, such an undertaking would likely provide a very useful extension of the GEM-PRO as a resource for such screens.
The limiting step of the overall approach is the SMAP runtime, which if implemented on a similar computing resource to that used in this study (see methods) would be limited to the order of hundreds of compounds screened against the E. coli GEM-PRO or tens of protein inhibitor screens against the ligand-bound PDB structures. Therefore, orders-of-magnitude more powerful computing resources would be necessary for massively parallel screens. This study builds upon previous examples [bib_ref] Drug off-target effects predicted using structural analysis in the context of a..., Chang [/bib_ref] [bib_ref] Structural systems biology evaluation of metabolic thermotolerance in Escherichia coli, Chang [/bib_ref] [bib_ref] Structure-based prediction of protein-protein interactions on a genome-wide scale, Zhang [/bib_ref] [bib_ref] Three-dimensional structural view of the central metabolic network of Thermotoga maritima, Zhang [/bib_ref] [bib_ref] Three-dimensional reconstruction of protein networks provides insight into human genetic disease, Wang [/bib_ref] [bib_ref] A structural systems biology approach for quantifying the systemic consequences of missense..., Cheng [/bib_ref] illustrating how structural and systems biology may combine to have an effect greater than they are capable of in isolation. For example, some of the SMAP predictions of lesser quantitative significance showed promise as antibacterial targets in simulation, sometimes accounting for known antibacterial targets that otherwise would have been called as false negatives by SMAP alone. Conversely, although metabolic model predictions have previously been shown to accurately predict the effects of many targeted gene knockouts [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref] and have been applied to select individual and multiple antibacterial targets [bib_ref] Targeting multiple targets in Pseudomonas aeruginosa PAO1 using flux balance analysis of..., Perumal [/bib_ref] [bib_ref] Comparative genome-scale metabolic reconstruction and flux balance analysis of multiple Staphylococcus aureus..., Lee [/bib_ref] , these metabolic models have not yet been capable of pairing these targets with compounds. Not only does the expansion from the GEM to GEM-PRO framework enable prediction of candidate compounds, it enables prediction of specific molecular mechanisms (e.g., competitive inhibition or complex disruption) that explain how the candidate compounds may affect the function of their targets.
In addition to providing a promising proof of principle that such a structural systems biology strategy can be used to understand antibacterial mechanisms, we have made specific predictions of chemical inhibitors of a protein currently unutilized for antibacterial applications (TrpB) and previously unknown mechanisms of existing antibacterial compounds, both those with and without established mechanisms. These predictions represent experimentally testable hypotheses and were generated entirely in silico. Therefore, Structural systems pharmacology may seed rapid discovery in the area of antibacterials.
# Conclusions
In this study, we developed an approach that can be used to predict and characterize antibacterial mechanisms either 1) by proteome-wide ligand binding target prediction and subsequent simulation of the effects of such interactions on growth or 2) by metabolic simulation of lethal protein loss of function and subsequent inhibitor prediction. This in silico approach bridges the gap between structural and systems pharmacology, linking molecular interactions with phenotypic outcomes. The GEM-PRO in this study enables proteome-wide binding site prediction specifically for E. coli metabolism, covering protein conformations in the physiological context of multimeric complexes including potential binding sites at proteinprotein interfaces. This is a foundational resource for antibacterial development for pathogenic E. coli and related species. The GEM-PRO was utilized to predict binding sites on protein targets for known antibacterials with unknown mechanisms (028 and 2OB), binding sites on previously uncharacterized targets of well-studied antibacterials (FCN and TOP), and potential inhibitors of TrpB. Furthermore, metabolic model simulations predicted specific essential processes by which these binding interactions would lead to antibacterial effects. These represent experimentally-testable hypotheses, and this study as a whole serves as a useful proof of principle for the structural systems pharmacology analysis of antibacterials.
# Methods
## Complex expansion of the e. coli gem-pro
Enzyme complexes included in the metabolic network iJO1366 [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref] were reviewed as annotated in EcoCyc [bib_ref] EcoCyc: a comprehensive database of Escherichia coli biology, Keseler [/bib_ref]. The annotation from EcoCyc includes protein subunit compositions, which served as a starting point for this reconstruction. The EcoCyc subunit compositions were evaluated from a structural perspective based on biological units of crystal structures in the PDB [bib_ref] The protein data bank, Berman [/bib_ref] and through thermodynamic analysis of possible physiological assemblies using the PDBePISA software [bib_ref] Inference of macromolecular assemblies from crystalline state, Krissinel [/bib_ref]. The most thermodynamically feasible PISA assembly for each complex, based on computed ΔG of dissociation, was compared to PDB biological units and EcoCyc composition annotation for each complex. In many cases, these three sources were in perfect agreement, in which case the PDB biological unit was chosen as the structure to represent the physiological assembly of the complex. However, many discrepancies were also found among the compositions assigned by these sources, including protein membership in complexes but missing stoichiometry in EcoCyc. To reconcile these discrepancies, the scientific literature was reviewed to find experimental evidence supporting the correct physiological assembly for a complex. These references reported data from a variety of experiments including: X-ray crystallography, gel filtration, size-exclusion chromatography, ultracentrifugation, functional assays, substrate binding assays, cooperative analysis, and mutant studies. A few studies also provided evidence from bioinformatics analysis such as kinetic assembly, molecular docking, and orthology-based inference. The consensus of these experimental results and the three preliminary sources was taken to determine the most likely physiological assembly. If the PDB biological unit agreed with the consensus, that structure was taken as the physiological assembly structure. If not, then the PISA structure that best agreed with the consensus was taken as the physiological assembly. In some cases, no PDB structure or PISA assembly completely accounted for the consensus complex assembly. In such cases, multiple structures were taken to represent as many sub-parts of the physiological complex assembly as possible. This resulted in some overlap with single-peptide chain structures included in the previously developed E. coli GEM-PRO.
## Smap implementation
SMAP was installed and run on a Linux server with 48core 1.9 GHz Opteron processor. For all results reported in this study, SMAP was run with default numerical parameters. The first SMAP run against a given query database and parameter set takes substantially longer than subsequent runs in order to define possible binding pockets (~55 h for the GEM-PRO and~629 h for all ligand-bound protein structures in the PDB). Average runtimes for subsequent screens in this study were~4 h and~49 h against the GEM-PRO and ligand-bound protein structures in the PDB, respectively.
## Protein-ligand interaction predictions
Different types of SMAP screens were run to answer three different types of questions: 1) positive and negative controls for antibacterials with known effective mechanisms in wild type E. coli K12 through known metabolic protein targets; 2) antibacterials known to be effective against E. coli K12 but with unknown mechanisms of action, seeking to answer the question of whether those compounds may target metabolic functions; 3) searches for potential novel antibacterials that are competitive inhibitors of metabolic proteins known to hinder growth of E. coli K12 if subjected to gene knockout. These are all open-ended questions, and candidate compounds and protein targets to be selected for these purposes are non-obvious. Also because SMAP is a method requiring substantial computational resources, the number of screens that could be performed was limited. For these reasons, filtering the wealth of candidate compounds and targets to choose candidates for the screens was necessary. Therefore, large data sources were filtered to pick most promising candidates to test these three types of questions.
## Selecting antibacterial controls for screen
As of September 24, 2012, there are 12,785 chemically distinct ligand molecules represented in at least one PDB structure. Given that SMAP performs best when starting with a well-defined ligand binding site for the search template, we chose only to use experimentally-determined binding sites for this type of screen. The collection of all known antibacterials and their known targets were collected from KEGG [bib_ref] KEGG: kyoto encyclopedia of genes and genomes, Kanehisa [/bib_ref] , EcoCyc [bib_ref] EcoCyc: a comprehensive database of Escherichia coli biology, Keseler [/bib_ref] , DrugBank [bib_ref] DrugBank 3.0: a comprehensive resource for 'omics' research on drugs, Knox [/bib_ref] , and ChEMBL [bib_ref] ChEMBL: a large-scale bioactivity database for drug discovery, Gaulton [/bib_ref] , and the overlapping set of these and the PDB ligands found. Antibiotic classifications were derived from KEGG, EcoCyc, and DrugBank. All PDB ligands were clustered by their chemical similarity using their canonical SMILESand the EI-Clustering software [bib_ref] Accelerated similarity searching and clustering of large compound sets by geometric embedding..., Cao [/bib_ref]. The distance matrix output by EI-Clustering was used to form the clusters by hierarchical clustering and a cutoff of 1.15 was determined such that the classified antibiotics were clustered together and not in the same clusters with antibiotics of other classes. Thus, functionally and chemically distinct groups of antibacterials were identified from which to choose positive controls. All curated data used for compound selection is presented in Additional file 4: . Positive controls were chosen from these groupings such that they represented a breadth of antibacterial classes and chemical clusters and only if they had at least one known metabolic protein target in E. coli.
Glucose was chosen as a negative control for this study due to multiple advantageous properties. Glucose is a molecule well known to cross the E. coli cellular membrane and not to exhibit negative effects on growth, as it is a primary carbon source for WT E. coli. Therefore, negative phenotypic effects would be completely unexpected in an accurate model. Glucose has many well-characterized binding sites, supported by a high number (> 400) of PDB structures in which it is co-crystalized with diverse proteins (representatives from > 200 protein clusters, with a 50% sequence identity threshold). Known binding targets for glucose in the E. coli GEM-PRO include five enzyme catalytic sites for which it is a known substrate and also as a competitive inhibitor of GlgP [bib_ref] Purification and properties of glycogen phosphorylase from Escherichia coli, Chen [/bib_ref] , providing test cases for ligand binding prediction as well as growth phenotype simulation upon target inhibition. As a small molecule (180 Da) within a standard deviation of the mean molecular mass of crystalized ligands in the PDB (376+/−196 Da), glucose is a reasonable representative of characterized ligands in terms of size. Glucose also satisfies Lipinski's rule of five [bib_ref] Experimental and computational approaches to estimate solubility and permeability in drug discovery..., Lipinski [/bib_ref] , indicative of its drug-like chemistry. These factors taken together make glucose a good negative control for all steps of our predictive approach.
## Selecting antibacterials with unknown mechanisms of action for screening
The ChEMBL database [bib_ref] ChEMBL: a large-scale bioactivity database for drug discovery, Gaulton [/bib_ref] was reviewed to find biological assays in which antibacterial activity of compounds was identified in E. coli. This set of compounds was searched for those with no known binding partners in WT E. coli according to KEGG, EcoCyc, DrugBank, ChEMBL, or the PDB. We then prioritized for those compounds that are ligands in PDB structures of only non-bacterial proteins. Small compounds consisting only of C, H, N, O, P, and S elements were chosen from this set as the orphan antibacterials of interest for this study. This data is also contained in Additional file 4: .
## Selecting orphan protein targets for screening
Previously published essentiality screens and simulations of the E. coli K12 single-gene knockout library grown on glucose minimal medium [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref] were analyzed to choose novel antibacterial protein targets to search for antimetabolites to inhibit. Phenotypes with very low growth at the end of the experiment (OD 600 < 0.26) were selected. Priority was given to proteins without known inhibitors in EcoCyc, DrugBank, or ChEMBL. From this set, three target proteins were chosen that bind to a high number of ligands in the PDB, have a low number of native metabolic substrates as annotated in iJO1366, and for which there is structural coverage in the GEM-PRO of the individual proteins, protein complexes, and catalytic sites. The curated data used for orphan protein target selection is presented in Additional file 5: .
## Prediction of antibacterial targets
In searching for possible metabolic protein targets for known antibacterial compounds, template structures were chosen from PDB crystal structures that included the compound bound to a protein. These structures were used with SMAP to search for potential binding pockets for these antibacterial compounds within both the previously published E. coli GEM-PRO and also the newly-generated physiological complex assemblies. The entire set of PDB proteins was clustered using a 50% sequence identity cutoff. The best resolution structure from each cluster that contained the ligand of interest was chosen as an alternative template for SMAP screens. SMAP was used to screen each template in turn across the database of proteins comprising the GEM-PRO structures. SMAP hits were considered significant for a p-value < 1.0 × 10 -4 and Tanimoto coefficient > 0.5. A secondary tier of lesser significance was determined using just the aforementioned p-value criterion.
## Prediction of anti-metabolite protein inhibitors
Searching for possible inhibitors of predicted antibacterial metabolic protein targets was performed by taking the structure of the protein target of interest from the E. coli GEM-PRO, docking [bib_ref] SwissDock, a protein-small molecule docking web service based on EADock DSS, Grosdidier [/bib_ref] the primary native metabolic substrate into the known catalytic site (as annotated in the GEM-PRO), and using the resulting structure as a template for SMAP screens. SMAP was then used to search across all ligand-bound protein structures in the PDB, excluding structures that only bind metal ions or metabolites included in iJO1366, to find ligands that bind to structurally similar sites. The query database contained 51,608 PDB structures. SMAP was run specifying that only ligand binding sites be considered. SMAP hits were considered significant with p-value < 1.0 × 10 -4 and Tanimoto coefficient > 0.5. A secondary tier of lesser significance was determined using just the aforementioned p-value criterion.
## Simulating protein inhibitory effects
The E. coli metabolic network iJO1366 [bib_ref] A comprehensive genome-scale reconstruction of Escherichia coli metabolism-2011, Orth [/bib_ref] was loaded into the COBRA toolboxfrom the published SBML model using Matlab. Since the time of publication of iJO1366 a thermodynamic constraint error was discovered in the published model; as a result, the malate oxidase, "MOX," reaction was set as irreversible. The superoxide dismutase, "SPODM," reaction was set with an initial upper bound of 1000 as well. The objective function was set as the complete wild type biomass reaction "Ec_biomass_iJO1366_WT_53p95M." Default exchange reaction constraints were used, except for a glucose uptake lower bound of −8 mmol/gDW/h and an oxygen uptake lower bound of −18.5 mmol/gDW/h, representing aerobic growth on glucose. These basic constraints were used for all reported simulations in this study.
The combined sets of known targets and predicted targets were first tested for antibacterial effects by constraining all associated reactions to 0 flux and then maximizing biomass using flux balance analysis (FBA) [bib_ref] What is flux balance analysis?, Orth [/bib_ref]. Individual targets were tested in the same manner to determine causal targets from the broader sets. Resulting biomass fluxes were compared to a simulated untreated condition where just the basic constraints were imposed and biomass was maximized; any decrease in biomass flux relative to the untreated condition was considered a prediction of antibacterial effect by degree of decrease.
## Analysis of impact of protein-ligand binding on molecular function
The specific amino acid residues comprising the ligand binding sites predicted by SMAP were compared to residue-resolution functional annotation contained in the original GEM-PRO [bib_ref] Structural systems biology evaluation of metabolic thermotolerance in Escherichia coli, Chang [/bib_ref]. If precise residues overlapped between these sets, we flagged these proteins as having predicted binding sites for the given ligand that should be seen as competitively inhibitory since they would bind to the same location as substrates required for normal function. Functional features included in this analysis consisted of catalytic sites and substrate binding sites. For SMAP query structures that were protein complexes containing multiple subunits, if the predicted ligand binding site included residues from distinct subunits, we flagged these as possible ligand binding events that could prevent or disrupt complex formation and therefore function.
## Additional files
Additional file 1: [fig_ref] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation... [/fig_ref]. Excel file containing indices providing details about the protein complex structures contained in the GEM-PRO. GEM-PRO file naming convention: (1) The PDB ID is given separated from the concatenated chain IDs by an underscore. [bib_ref] Detecting evolutionary relationships across existing fold space, using sequence order-independent profile-profile alignments, Xie [/bib_ref] The stoichiometric presence of each chain is annotated in parentheses following each chain ID. No parenthetic number indicates a stoichiometry of 1. (3) Suffixes for PDB biological units are retained, such as ".pdb1", ".pdb2", or ".pdb3". (4) The suffixes ".pisa1.pdb", ".pisa2.pdb", ".pisa3.pdb", and ".pisa4.pdb", indicate that the structure is an output of PDBePISA. The number following ".pisa" indicates the rank within the list of possible structures returned by PDBePISA.
Additional file 2: [fig_ref] Table 2: Metabolic model performance in predicting antibacterial effects [/fig_ref]. Excel file containing summary results from all SMAP screens.
Additional file 3: [fig_ref] Figure 1: phan synthase β subunit [/fig_ref]. Sub-networks of iJO1366 affected by simulated inhibition of predicted targets of 028, 2OB, and TrpB inhibition by F6F, PLT, 7MN, IDM, or PLS. Reactions in green, red, and yellow are those directly affected by predicted target inhibition by 028, 2OB, and one of the predicted TrpB inhibitors, respectively. Reactions with thicker lines represent those with lower magnitude flux upon simulated exposure to these
[fig] Figure 1: phan synthase β subunit (TrpB). Predicted TrpB inhibitors include 2-{[4-(trifluoromethoxy)benzoyl]amino} Complex expansion of E. coli GEM-PRO. (A) This expansion of the E. coli GEM-PRO provides structural coverage of protein complexes included in iJO1366. An example is depicted for the GlmU protein catalyzing the G1PACT reaction. (B) Complete and partial coverage of each protein complex by at least one structure is categorized. (C) The oligomeric states of complexes for which there is complete coverage in this GEM-PRO are distributed across monomers, homomultimers, and heteromultimers. ethyl dihydrogen phosphate (F6F), [3-hydroxy-2-methyl-5phosphonooxymethyl-pyridin-4-ylmethyl]-L-ryptophane (PLT), (Z)-N-[(1E)-1-carboxy-2-(2,3-dihydro-1H-indol-1-yl) ethylidene]{3-hydroxy-2-methyl-5-[(phosphonooxy) methyl]pyridin-4(1H)-ylidene}methanaminium (7MN), indoline (IDM), and pyridoxyl-serine-5-monophosphate (PLS). Criteria supporting the potential inhibitors of TrpB are listed in [/fig]
[fig] Figure 3: SMAP performance in recalling true positives. The lowest rank for each protein structure predicted as an SMAP hit is displayed for the set of known protein targets for the five control compounds. Blue lines indicate the rank position (out of 3237) of a known target for a given compound. n = the number of screens using different protein structure templates performed for each compound. p = the p-value resulting from Mann Whitney statistical tests for individual SMAP results with respect to an individual template screen. BGC: beta-D-glucose; FCN: fosfomycin; YTZ: 4-amino-N-(1,3-thiazol-2-yl)benzenesulfonamide; TOP: trimethoprim; H3P: 2,2′-methanediylbis(3,4,6-trichlorophenol). [/fig]
[fig] Figure 4: Predicted antibacterial mechanisms. (A) Inhibition of predicted binding targets (IspA and IspB) of 028 impacted simulated growth through decreased flux through isoprenoid synthesis pathways, leading to no growth under complete inhibition. (B) Through simulated inhibition of predicted binding targets of 2OB, critical metabolic pathways were impacted leading to decreased growth: PheA impacting amino acid synthesis, AcpP impacting lipid synthesis, EntA impacting enterochelin metabolism, and AtpB, cytochrome bo terminal oxidase, and succinate dehydrogenase all impacting oxidative phosphorylation. (C) Five compounds (F6F, PLT, 7MN, IDM, and PLS) were predicted to bind and competitively inhibit TrpB, leading to decreased tryptophan synthesis. [/fig]
[table] Table 1: Summary of in silico antibacterial predictionsNext, we utilized the residue-resolution functional annotation of the previously generated E. coli GEM-PRO to identify whether the SMAP-predicted ligand binding sites overlapped with known functional sites, such as catalytic and substrate binding sites. Such interactions could be expected to exhibit competitive inhibitory effects. For cases where an SMAP prediction was made on the basis of a protein complex structure, we also identified predicted ligand binding sites at the interface between subunits, which may lead to disruption or prevention of protein complex formation in vivo and therefore have a deleterious impact on enzyme function. Overlap between predicted TOP binding sites and native nucleotide and substrate binding sites occurred on RibD, partial overlap with the catalytic site of IspU, and almost complete overlap with the catalytic sites of both EntA and FabG. The predicted binding sites for 028 completely overlapped with the catalytic site of IspA and overlapped with the substrate binding site and Mg 2+ ion binding site of IspB. In the case of 2OB, predicted binding sites showed at least partial overlap with the catalytic sites of PheA, CyoB, EntA, AtpB, and AcpP. Predicted 2OB binding sites also had [/table]
[table] Table 2: Metabolic model performance in predicting antibacterial effects [/table]
|
HIV-1 Protease and Reverse Transcriptase Control the Architecture of Their Nucleocapsid Partner
## Supporting information
Animation S1. An animated model of HIV-1 reverse transcription. In order to better understand the overall process of HIV reverse transcription, an animated model is available here for downloading. Animation S1 is a MOV file for viewing with Apple Quicktime and Animation S2 is a SWF file for viewing with the . For clarity, this animation represents an idealized process with a figure-ofeight-shaped nucleic-acid template. Attention has been focused on the following points: the three primers (tRNALys3, 39ppt and cppt) are oversized for clarity; the two RNA ends (RU5 and U3R) are in close proximity throughout reverse transcription, in order to show a continuous process at the time of the strand transfers; the RTassociated RNAseH is shown with a high activity for digestion of the RNA following minus strand DNA elongation. The two polypurine tracts are presented as the two exclusive primers for plus strand DNA initiation, whereas the remaining RNA fragments are proposed to be displaced by the two translocating RT molecules during minus strand DNA elongation. The central DNA flap is viewed as a fluctuating structure, the function of which still needs to be elucidated. The presentation of the timing for plus strand DNA termination in combination with the central flap synthesis and LTR duplication seems the most representative to the authors. This model does not exclude alternative pathways, especially considering the dimeric nature of the initial RNA template. (0.13 MB MOV) (MOV) Animation S2. See Legend of Animation S1. (0.13 MB SWF) (SWF) |
Contribution of Home and School Environment in Children’s Food Choice and Overweight/Obesity Prevalence in African Context: Evidence for Creating Enabling Healthful Food Environment
This review aimed primarily to investigate the current trends of overweight and obesity in school children in the African context, secondly to explore the contribution of home and school environments on the children's food choices and lastly suggesting measures for creating a healthier food environment. Despite the increase in overweight and obesity among school children, empirical evidence on their determinants in the African context is scarce, thus calls for consideration of home and school environments. A literature search was conducted between October and December 2018 using Medline (PubMed), Directory of Open Access Journals, Google Scholar, manual search and "grey" literature. This review included articles published between the 1st January 2008 and 30th June 2018. Out of 343 articles, 49 were included for the full text reading after meeting the inclusion criteria. Five reports from grey literature were also included. Results show that the prevalence of overweight and obesity among school children in Africa is increasing and ranges from <5% to >40% in the 10-year period in which the review was taken. High socio-economic status, urban residence and female gender predicted higher prevalence of overweight/obesity. Few reviewed articles on the contribution of home and school environments on children's food choices showed a shred of evidence, thus calls for further research to address this gap. This review found an increasing prevalence of overweight and obesity in school children in Africa. Therefore, further investigation of home and school environment is imperative to curb the increase in the magnitude of overweight and obesity.
# Introduction
In recent decades fast increase in the prevalence of obesity and non-communicable diseases is evident in urban areas of low and middle-income countries. [bib_ref] Overweight and obesity in children and adolescents: the South African problem, Rossouw [/bib_ref] [bib_ref] Association between home and school food environments and dietary patterns among 9-11-year-old..., Vepsäläinen [/bib_ref] [bib_ref] Prevalence of overweight and obesity among primary school children in a developing..., Pienaar [/bib_ref] [bib_ref] Risk factors for obesity and overfat among primary school children in Mashonaland..., Kambondo [/bib_ref] Overweight and/or obesity is referred as excess accumulation of fat in the body which is a precursor for non-communicable diseases. [bib_ref] Prevalence of childhood overweight/ obesity in basic school in Accra, Mohammed [/bib_ref] [bib_ref] Prevalence of overweight, obesity, and thinness in Cameroon urban children and adolescents, Wamba [/bib_ref] [bib_ref] Overweight/obesity among school aged children in Bahir Dar City: cross sectional study, Mekonnen [/bib_ref] According to the International Obesity Task Force overweight in children is defined as a value between 85th and 95th percentile, while obesity is a value above 95th percentile. [bib_ref] Comparison between international obesity task force and centre for disease control references..., Hajian-Tilaki [/bib_ref] World Health Organization has defined childhood overweight , as a value of +1 standard deviation (SD) which is equivalent to 25 kg/m 2 cut-off for adults and a value of +2 standard deviation kg/m 2 ) as obesity which is close to adult's cut-off of ≥ 30.0 kg/m 2 ) [bib_ref] Childhood Obesity Estimates Based on WHO and IOTF Reference Values, Alqahtan [/bib_ref] Globally, there are about 42 million overweight children with over 35 million living in developing countries. [bib_ref] Childhood obesity: causes and consequences, Sahoo [/bib_ref] In Africa, despite high levels of undernutrition, overweight and obesity rates in children are increasing, [bib_ref] Overweight/obesity among school aged children in Bahir Dar City: cross sectional study, Mekonnen [/bib_ref] a pattern often attributed to limited policy support and lack of multi-sectoral collaboration. [bib_ref] Prevalence of overweight and obesity among primary school children in a developing..., Pienaar [/bib_ref] The 2018 Global Nutrition Report, showed that 4.6% of boys and 4.7% of girls were obese globally. In Africa, for adolescent aged 10-19 years 2.5% of boys and 3.9% of girls were reported as obese in 2016.In sub-Saharan Africa, the prevalence of overweight and obesity is more than 10% in many countries such as; South Africa, Ethiopia, Cameroon, Nigeria, Kenya, and Tanzania (see [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. Several studies have reported a prevalence of more than 20% in childhood obesity across African nations [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. A prevalence of more than 30% was reported in Uganda, and Egypt, whereas some studies in Zimbabwe, Tanzania and Sudan found a prevalence of less than 10% (see [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. Overweight and obesity in children have devastating physiological consequences, namely high cholesterol, and high blood pressure, high risk of impaired glucose tolerance, insulin resistance, type 2 diabetes, breathing problems and fatty liver disease, [bib_ref] Consequences of childhood obesity, Lee [/bib_ref] as well as psychosocial consequences, such as stigma, teasing, and harassment. [bib_ref] Overweight and obesity in children and adolescents: the South African problem, Rossouw [/bib_ref] [bib_ref] Psychological consequences of childhood obesity: psychiatric comorbidity and prevention, Rankin [/bib_ref] The food environment across countries is in a state of rapid transformation due to changes in dietary patterns and food behaviors. [bib_ref] Measures of the food environment, a compilation of the literature, Mckinnon [/bib_ref] [bib_ref] Prevalence and factors associated with overweight and obesity among children from primary..., Sagbo [/bib_ref] The food environment is rooted in the recognition that an individual is surrounded by broad dimensions of obtaining and consuming food. It encompasses the extent to which someone accesses available food which is desirable and convenient, the way food is marketed, and exclusive properties of food, such as taste and appearance. [bib_ref] The local food environment and diet: A systematic review, Caspi [/bib_ref] Processed, convenient, and relatively low price foods are now readily available in many settings. [bib_ref] Creating healthy food and eating environments: policy and environmental approaches, Story [/bib_ref] Some studies [bib_ref] Association between home and school food environments and dietary patterns among 9-11-year-old..., Vepsäläinen [/bib_ref] [bib_ref] Childhood obesity: determinants, evaluation, and prevention, Raychaudhuri [/bib_ref] [bib_ref] Teachers' perceptions of school nutrition education's influence on eating behaviours of learners..., Kupolati [/bib_ref] reported that the school food environment contributes to children's food choices. Along with strong peer pressure, children feel forced to buy and consume these unhealthy food alternatives. [bib_ref] Teachers' perceptions of school nutrition education's influence on eating behaviours of learners..., Kupolati [/bib_ref] School children and adolescents require adequate nutrient and energy inputs for growth, development, and good academic performance. [bib_ref] Factors influencing the food choices of Irish children and adolescents: A qualitative..., Fitzgerald [/bib_ref] [bib_ref] The dual burden of malnutrition and associated dietary and lifestyle habits among..., El-Kassas [/bib_ref] Appropriate dietary choices and dietary intake are crucial for building good eating habits early in life. [bib_ref] Factors influencing the food choices of Irish children and adolescents: A qualitative..., Fitzgerald [/bib_ref] [bib_ref] Dietary intake of schoolchildren and adolescents in developing countries, Ochola [/bib_ref] Eating habits during childhood may have long term implications; therefore, environments associated with children's food choices need to be studied for modification/reinforcement of healthy choices. Foods that are high in sugar and/or fat, which are globally available in school environments are associated with increased risk of obesity [bib_ref] The school food environment and student BMI and food consumption: 2004 to..., Terry-Mcelrath [/bib_ref] and their consumption is reported to increase at the highest rates ever. [bib_ref] Relationship between soft drink consumption and obesity in 9-11 years old children..., Katzmarzyk [/bib_ref] [bib_ref] Differential prevalence and associations of overweight and obesity by gender and population..., Negash [/bib_ref] Escalating overweight and obesity rates in school children necessitate consideration of both home and school food environments as potential contributors. 14,26 Children spend significant amounts of time in the school environment which may influence their food choices and shape their attitudes towards foods offered in the school. [bib_ref] Primary school food environment in Mauritius, Sun [/bib_ref] The home environment (through the physical presence of food and television advertisement) also influences children's food choices. [bib_ref] Exposure to 'healthy' fast food meal bundles in television advertisements promotes liking..., Boyland [/bib_ref] Couch and colleagues [bib_ref] Home food environment in relation to children's diet quality and weight status, Couch [/bib_ref] reported that availability of unhealthy foods in the home has been associated with a lower intake of fruits and vegetables. Wang et al [bib_ref] Home food environment, dietary intake, and weight among overweight and obese children..., Wang [/bib_ref] found that the home environment is responsible for shaping children's lifelong eating habits. At home, parents play a key role in children's exposure to foods. [bib_ref] Parents' food choice motives and their associations with children's food preferences, Russell [/bib_ref] To the best of our knowledge, no review of the existing literature has investigated obesity prevalence and contributions of school and home environments in children's food choices in the African context. Therefore, the objectives of this review article were (i) to review the existing literature on current trends of overweight and obesity in Africa (ii) to determine the contribution of home and school environments to children's food choices and (iii) to suggest strategies for creating enabling healthy food environments for children.
# Methods
## Search strategy, study selection
A systematic literature search was conducted for articles published on the prevalence of overweight and obesity in school children in Africa. Other articles searched were on the contribution of home and school environment in children's food choices. In addition, reports constituting "grey literature" on strategies for improving children's food environments were searched. A comprehensive keyword search was used for terms related to the topic of interest. We included original published articles conducted in Africa, human studies, reference age group (6-18 years) and within 10 years of publication. Studies done outside Africa, review articles and those which did not use the English language were excluded. A manual search via Google was also conducted from the reference lists of reviewed articles.
Three databases were included in this review: Medline (PubMed), Directory of Open Access Journals (DOAJ),
## Data extraction
Major findings/data on the prevalence of overweight and obesity in African school children are summarized in [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. Major findings of the articles on influence of home and school food environment on food choice were summarized in [fig_ref] Table 2: Summary of Studies on Influence of School and Home Environment on Children... [/fig_ref] , with the following subheadings: name of the first author and year of publication, country, environment studied (home or school), study design and description of main findings of the study. Information from grey literature was also included in the description of studies.
# Results
## Description of included studies
Of the 343 articles identified 294 were excluded, finally, 49 studies and 5 reports "grey literature" were included in the review [fig_ref] Figure 1: Flow chart of the selection process for inclusion of eligible studies [/fig_ref]. About 39 studies investigated the prevalence of overweight and obesity in African countries, and 10 articles reported on the influence of home and school food environment on school children's food choices. Grey literature provided insights on strategies for creating enabling healthy food environments except Global Nutrition report which provided data on the global prevalence of overweight and obesity in children and adolescents. Other reports were from World Health Organization (WHO), Food and Agriculture Organization (FAO) Tanzanian Ministry of Health, Global Agricultural Information Network. Thirty-six of the reviewed articles on overweight and obesity prevalence used a crosssectional study design, two studies used the analysis of secondary data and one study used longitudinal design. Details are presented in [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. Three out of ten articles [fig_ref] Table 2: Summary of Studies on Influence of School and Home Environment on Children... [/fig_ref] on school and home environment used qualitative methods, four studies used a cross-sectional method, one of each article used a mixed-methods approach, a cohort and a quasi-experiment.
## Key findings of this review
There is observed increase in overweight and obesity prevalence between studies and countries although there are notable variations.
Northern and eastern African regions had higher rates of overweight and obesity compared to southern and horn of Africa regions.
Most of the studies were cross-sectional and reported higher prevalence of overweight and obesity in girls than in boys.
Different cut-off values (WHO, IOTF, CDC) were used to define overweight and obesity which may cause variation in prevalence. Therefore interpretation of results needs a caution.
There is limited information on contribution of home and school food environment in children's food choice as well as link between food choice and obesity prevalence. Therefore, creating a research gap in this area.
Proposed strategies would shed some lights on creating enabling healthful food environment for school children in Africa.
## Overweight and obesity prevalence among school children in africa
Evidence from reviewed articles [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref] showed diverse trends of overweight and obesity among school children across African countries. In South Africa, prevalence varied considerably between 5% to > 20% across studies in different years. Studies reported in 2010 and 2011 showed a low prevalence than studies reported from 2014. Studies by [bib_ref] Obesity in 7-10-year-old children in urban primary schools in Port Elizabeth, Baard [/bib_ref] [bib_ref] Differential prevalence and associations of overweight and obesity by gender and population..., Negash [/bib_ref] [bib_ref] Relationships between overweight, obesity and physical fitness of nine-to twelve-year-old South African..., Truter [/bib_ref] reported that girls were likely to be more obese than boys [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. In Nigeria the combined prevalence for overweight and obesity ranged between 2.8% in 2013 36 to 12.2% in 2018. [bib_ref] Prevalence of overweight and obesity among school-age children in Jos, Ofakunrin [/bib_ref] The prevalence of overweight and obesity was higher in children than in adolescents, 5 also higher in girls than in boys. [bib_ref] Prevalence of overweight and obesity among school-age children in Jos, Ofakunrin [/bib_ref] In Ghana, a study by Annan-Asare and colleagues 39 found a higher prevalence in both girls and boys than earlier work of 6 which reported lower rates of overweight and obesity, although girls were 2 times more likely to be obese than boys. In Cameroon, [bib_ref] Prevalence of overweight, obesity, and thinness in Cameroon urban children and adolescents, Wamba [/bib_ref] and Zimbabwe, 4 the prevalence of overweight and obesity in children was < 15%. A Low prevalence of overweight and obesity (< 6%) was reported in Togo. [bib_ref] Prevalence and factors associated with overweight and obesity among children from primary..., Sagbo [/bib_ref] In Lesotho, one study reported that girls were found to be more overweight than boys. [bib_ref] Body mass index of 16-year olds in urban Maseru, Van Den Berg [/bib_ref] In Egypt the prevalence rate of overweight and obesity ranged between 31->40%. In one study 42 girls were more overweight and obese than boys while other studies [bib_ref] Prevalence of overweight and obesity among preparatory school adolescents in Urban Sharkia..., Talat [/bib_ref] [bib_ref] Prevalence of overweight and obesity in primary school children in Port Said..., Badawi [/bib_ref] showed slight variations. In Libya, two studies showed a prevalence range of 9.5-28.2%. Boys were more overweight and obese than girls [bib_ref] Perspective on childhood obesity in general and libya in particular, Sheriff [/bib_ref] and in the other study [bib_ref] Overweight and obesity among primary school children in misurata city-libya: prevalence and..., Hussein [/bib_ref] the prevalence of overweight and obesity ranged between 7.7-25.3%, the prevalence was low in recent study [bib_ref] Prevalence of obesity, overweight, underweight and stunting among school children in Argo..., Hussein [/bib_ref] compared to previous studies [bib_ref] Alarming high prevalence of overweight/obesity among Sudanese children, Nagwa [/bib_ref] which was related to the high socioeconomic status of parents. [bib_ref] Alarming high prevalence of overweight/obesity among Sudanese children, Nagwa [/bib_ref] In Ethiopia overweight and obesity prevalence ranged between 11.9-20.5%. In these studies, boys were more obese than girls, but girls were more overweight. Moreover, in private school prevalence was significantly higher than in public schools. [bib_ref] Childhood overweight, obesity and associated factors among primary school children in dire..., Desalew [/bib_ref] Surprisingly, a study done in 2018 showed a lower prevalence than a study reported in 2017.
In East Africa, evidence shows that childhood overweight and obesity trends have increased at an alarming rate. In Kenya, studies of 51-53 found a high prevalence of overweight and/or obesity in which girls were more likely to be overweight/obese than boys and high in private schools than public schools, 52 the prevalence range between these studies was 17.8-20.8% and there is slight variation over time. In Uganda, a prevalence of >50% (overweight and obesity combined) was reported by. [bib_ref] Prevalence of overweight and obesity among primary school children in Kampala central, Chebet [/bib_ref] In a national survey of two countries (Uganda and Ghana), the prevalence of overweight and obesity in girls was significantly higher than in boys. [bib_ref] Overweight and obesity and associated factors among school-aged adolescents in Ghana and..., Peltzer [/bib_ref] Tanzania, like other East African countries, is not immune to the growing problem of overweight and obesity in school-children as reflected by studies done across the country (see [fig_ref] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children... [/fig_ref]. The prevalence range was between 9.2-22.6%. Studies by [bib_ref] Prevalence and determinants of obesity among school children in Dar es Salaam, Muhihi [/bib_ref] [bib_ref] Prevalence of overweight and obesity among primary school children aged 8-13 in, Pangani [/bib_ref] found that girls were more overweight and obese than boys, and is higher in private schools than in public schools. [bib_ref] Prevalence of overweight and obesity among primary school children aged 8-13 in, Pangani [/bib_ref] Also, a low prevalence was found in a study [bib_ref] Magnitude and factors associated with overweight and obesity among adolescents in semi-rural..., Tluway [/bib_ref] done in a rural area.
## Influence of school and home environment
Ten African-based studies (six from the school environment and four from both school and home environment) investigated the contributions of school and home environments on children's food choices and dietary habits (see [fig_ref] Table 2: Summary of Studies on Influence of School and Home Environment on Children... [/fig_ref]. Three studies 20,65,66 revealed that perception and availability of healthy food options in the home and school environments can improve children's food choices. One study 67 associated BMI with food consumption. It reported that children who brought lunch boxes to school had lower BMI than children who consumed foods purchased from school tuck shops. These shops are characterized by high energy-dense foods. Other studies [bib_ref] Primary school food environment in Mauritius, Sun [/bib_ref] [bib_ref] Consumption of sweetened beverages among school-going children in a densely populated township..., Kalimbira [/bib_ref] [bib_ref] Added sugar and dietary sodium intake from purchased fast food, confectionery, sweetened..., Feeley [/bib_ref] revealed the presence of unhealthy food options in either home or school environment which promotes consumption of these foods by school children. Parents and peers were also found to dictate unhealthy food choices by school children. [bib_ref] Perceptions and attitudes towards food choice in adolescents in Gaborone, Brown [/bib_ref]
# Discussion
## Overweight and obesity prevalence in school children
Obesity has now reached epidemic levels in both developing and developed countries. [bib_ref] Prevalence of overweight and obesity among children aged 6-12 years in Dodoma..., Mosha [/bib_ref] [bib_ref] Prevalence of overweight and obesity among primary school children aged 8-13 in, Pangani [/bib_ref] Different studies in Africa have shown the increased prevalence in obesity from the lowest reported at 5% to the highest at above 40% across the review period of 10 years. Only 4 studies reported a prevalence of less than 10% when overweight and obesity are combined and 35 studies combined prevalence was more than 10%. This trend is almost similar to what was reported in developed countries and some other developing countries in the global south. For instance, in Latin America, more than 20% of schoolchildren were obese [bib_ref] Nutrition status of children in Latin America, Corvalán [/bib_ref] and in San Diego, 26.6% of children were overweight and obese. [bib_ref] Home food environment in relation to children's diet quality and weight status, Couch [/bib_ref] Similarly, high prevalence of overweight and obesity (39.7%) were found in Mexican school children [bib_ref] Caloric beverage consumption patterns in Mexican children, Barquera [/bib_ref] while 31% of Canadian, [bib_ref] Examining local-level factors shaping school nutrition policy implementation in Ontario, Vine [/bib_ref] [bib_ref] Overweight and Obesity among Adolescents in Kano, Yusuf [/bib_ref].6% of Saudi Arabian 73 and 32% of American 30 school children were reported to be overweight and obese. This review observed variation in overweight and obesity prevalence between studies and over time. Therefore, data on overweight and obesity are complex and confusing, this may be due to different contexts and age groups under investigation. Some evidence showed that cultural and environmental factors may explain variation in obesity data among study subjects. [bib_ref] Prevalence of childhood and adolescent overweight and obesity in Kano State, Umar [/bib_ref] [bib_ref] Prevalence of overweight and obesity among primary school children aged 8-13 in, Pangani [/bib_ref] Cultural norms in Africa favor large/fat body sizes especially in girls and women, thus it could be a barrier to maintain healthy body weight. [bib_ref] Overweight and obesity among women: analysis of demographic and health survey data..., Neupane [/bib_ref] The difference in socio-economic status also contributed to these variations, as children from higher Social Economic Status are reported to be more obese than children from lower SES. [bib_ref] Overweight/obesity among school aged children in Bahir Dar City: cross sectional study, Mekonnen [/bib_ref] [bib_ref] Obesity in 7-10-year-old children in urban primary schools in Port Elizabeth, Baard [/bib_ref] [bib_ref] Prevalence of childhood and adolescent overweight and obesity in Kano State, Umar [/bib_ref] [bib_ref] Alarming high prevalence of overweight/obesity among Sudanese children, Nagwa [/bib_ref] Another variation may be attributed to different cut-off values used to define overweight and obesity and study designs. [bib_ref] Prevalence and factors associated with overweight and obesity among children from primary..., Sagbo [/bib_ref] For example, in a study with Iranian adolescent CDC reference values obtained a higher rate of obesity by 3.2% than IOTF reference values in children aged 12-15 years of both sexes. [bib_ref] Comparison between international obesity task force and centre for disease control references..., Hajian-Tilaki [/bib_ref] Similarly, in Saudi Arabian adolescents WHO reference values showed a higher rate of obesity (almost by 6%), but lower rates of overweight (almost by 5%) than IOTF. [bib_ref] Childhood Obesity Estimates Based on WHO and IOTF Reference Values, Alqahtan [/bib_ref] Children from urban areas were found to be more obese 43,75 than children from rural areas. [bib_ref] Magnitude and factors associated with overweight and obesity among adolescents in semi-rural..., Tluway [/bib_ref] [bib_ref] Overweight, obesity, underweight and stunting in female primary school learners in rural, Tathiah [/bib_ref] Similarly,
## Dovepress
Zhang et al 2016 [bib_ref] Patterns and determinants of double-burden of malnutrition among rural children: evidence from..., Zhang [/bib_ref] found an increased risk of overweight and obesity in urban Chinese school children, which may be caused by a difference in lifestyle behaviors between rural and urban settings. Generally, most urban populations have sedentary lifestyles and increased consumption of high energy-dense foods. [bib_ref] Prevalence and determinants of obesity among school children in Dar es Salaam, Muhihi [/bib_ref] To great extent urbanization has greatly affected African food culture and resulted in a shift in dietary consumption 97 which leads to increased consumption of westernized diets which are perceived as a symbol of affluence. [bib_ref] Overweight and obesity among women: analysis of demographic and health survey data..., Neupane [/bib_ref] Along with urbanization is characterized by motorized lifestyle and physical inactivity. [bib_ref] Overweight and obesity among women: analysis of demographic and health survey data..., Neupane [/bib_ref] Many studies across countries reported a higher prevalence of overweight and obesity in girls compared to boys. This may be attributed to differences in gender roles where boys are more active than girls. [bib_ref] Prevalence of childhood and adolescent overweight and obesity in Kano State, Umar [/bib_ref] Boys spent adequate time in fieldwork while girls spend more time at home. [bib_ref] Childhood overweight, obesity and associated factors among primary school children in dire..., Desalew [/bib_ref] Early-onset of menarche in girls is associated with an increase in body fat and body weight. [bib_ref] Magnitude and factors associated with overweight and obesity among adolescents in semi-rural..., Tluway [/bib_ref] Also, children below 10 years were more likely to be obese than children/adolescents above 10 years. [bib_ref] Prevalence of overweight and obesity in primary school children in Port Said..., Badawi [/bib_ref] [bib_ref] Prevalence of overweight and obesity among children aged 6-12 years in Dodoma..., Mosha [/bib_ref] This pattern may be attributable to faulty feeding during early childhood. Children from private schools were associated with higher obesity rates, and the possible explanation for this was, private school children were linked to high social-economic status thus easily adopting unhealthy food choices. [bib_ref] Childhood overweight, obesity and associated factors among primary school children in dire..., Desalew [/bib_ref] [bib_ref] Feeding Habits associated with overweight and obesity amongst secondary school students in..., Rapando [/bib_ref]
## Home and school food environment
There is limited information on the influence of home and school environments on school children's food choices and obesity linkages. However, reviewed studies have demonstrated that obesity is a significant health problem and provide direction for evidence-based strategies and interventions. This study found that positive perception in healthy foods would improve children's healthy food choices. This was also revealed by a study in Southern Appalachia that strict parental control have influence in healthy food choices by children regardless of the availability of unhealthy food alternatives in the home environment, 30 this mighty influence healthy food choices during adolescence and adulthood. Some of the reviewed studies also found that unhealthy food options are readily available in African school environments as well as in the home environment, and children prefer to choose and consume these foods. These findings are similar to studies from other settings, Brazil, [bib_ref] Nutrition status of children in Latin America, Corvalán [/bib_ref] South Africa, [bib_ref] The school food environment: shaping the future health of the nation, Villiers [/bib_ref] India, [bib_ref] Childhood obesity: determinants, evaluation, and prevention, Raychaudhuri [/bib_ref] San Diego. [bib_ref] Home food environment in relation to children's diet quality and weight status, Couch [/bib_ref] Only one study from South Africa linked BMI and food intake. This is in line with a study from Saudi Arabian school children which found a positive association between increased intakes of sugar-sweetened carbonated beverages with an increase in BMI. [bib_ref] Sugar-sweetened carbonated beverage consumption correlates with BMI, waist circumference, and poor dietary..., Collison [/bib_ref] Studies among Irish children [bib_ref] Factors influencing the food choices of Irish children and adolescents: A qualitative..., Fitzgerald [/bib_ref] and Vancouver Canada adolescents [bib_ref] The home food environment and associations with dietary intake among adolescents presenting..., Watts [/bib_ref] revealed that parents had the most significant control on child's food intake. In addition, parents have more influence on portion sizes to be offered to children, however, decision, motivation and parental feeding goals are not well understood. [bib_ref] Determinants of children's eating behavior, Scaglioni [/bib_ref] Presence of parents at home and their involvement in feeding practices influence role modeling, shape positive behavior change and modify available foods in the home environment. [bib_ref] Influence of parenting practices on eating behaviors of early adolescents during independent..., Reicks [/bib_ref]. In addition to their presence at home, parents also are involved in family food preference, time allocation, prioritization of activities, preparation of food, skills, financial and health attributes. Children also are imitating these attributes. [bib_ref] Children's food choice process in the home environment. A qualitative descriptive study, Holsten [/bib_ref] A study in Jordan revealed that peer influence associated significantly with disordered eating behavior among adolescents with respect to their body images. [bib_ref] Relationship between peer pressure and risk of eating disorders among adolescents in..., Al-Sheyab [/bib_ref] A recent systematic review of literature revealed that peers and siblings have both positive (healthy eating) and negative (unhealthy eating) influence on food choices of children. However, negative influence is most common. 100
## Strategies for creating enabling healthful food environment
## Conducting informative research
Effective implementation of optimal intervention strategies to prevent obesity in the African context needs sufficient evidence on current and periodic trends of obesity across countries. [bib_ref] Prevalence of obesity and overweight in African learners: A protocol for systematic..., Adom [/bib_ref] Knowledge about food choice and dietary habits data among African school children is inadequate, therefore, more consumer research on the school and home food environment are required. Studies in developing countries need to examine the role of local/informal food vendors and other sources of food like production and food donations on food choice.There is a need to focus on the penetration of supermarkets that offer highly processed food varieties.Nevertheless, before launching school-based interventions in low and middle-income countries, we need to establish culturally based evidence. This is because of differences in values, norms, customs, and environmental influences on food choices for children. The existing models from developed countries may not work in the African context. [bib_ref] A conceptual framework for healthy eating behavior in Ecuadorian adolescents: A qualitative..., Verstraeten [/bib_ref] Barriers and facilitators influencing healthy eating behaviors in the home and school environment need to be thoroughly investigated as
## Designing social-ecological frameworks for africa
Apart from biophysical factors (genes, age, gender) the ecological perspective states that the physical environment has a direct link to obesity due to its influence on food choices. The social-ecological perspective describes relationships between an individual and environment positing that individual behavior emerges from the interplay of multiple factors between the two entities. [bib_ref] Childhood obesity in developing countries: epidemiology, determinants and prevention, Gupta [/bib_ref] The social-ecological framework considers five levels of influence: Individual (knowledge, attitude, and self-concept); Interpersonal (family, peers, friends, social networks); Community (relationships between organizations); Organizational (organizations and social institutions); and Policy/enabling environment (national, state, local). At the individual level, school children need to receive nutrition education to equip them with appropriate knowledge to make informed healthy food choices and change their attitudes and skills to build self-efficacy. [bib_ref] Dietary habits and eating practices and their association with overweight and obesity..., Sedibe [/bib_ref] [bib_ref] Nutrition education tools for primary school children in the Vaal region, Oldewage-Theron [/bib_ref] At the interpersonal/ family level, training of parents and modification of the home environment is required. It is also imperative to create awareness through health promotion campaigns targeting parents, households, and communities on a healthy diet and dietary diversity. Dietary diversity should promote the consumption of nutrient-dense culturally acceptable foods and minimize the consumption of processed foods. [bib_ref] Temporal changes and determinants of childhood nutritional status in Kenya and Zambia, Hoffman [/bib_ref] African communities should maximize development of diets that uses local food staffs. Parents have the role to model healthy eating habits and food choices to their children. Parents should be well informed on the outcome of poor dietary habits of their children and be motivated to become good examples. [bib_ref] Determinants of children's eating behavior, Scaglioni [/bib_ref] At the organization/school level, teachers are capacitated to deliver nutrition education and communication messages to pupils. Promotion/modification of school food policy/environment and creating resources for physical activity can easily be designed and implemented at the school level. National and local levels are obliged to set policies and laws that will promote the creation of a healthful enabling food environment. The development of ecological framework need to consider African culture. Socio-cultural beliefs has influence in nutrition matters of families since some foods are highly prized to complete a meal and these foods may have influence on weight changes. Some food taboos are related to particular foods, for example foods from animal origin. [bib_ref] Childhood obesity: causes and consequences, Sahoo [/bib_ref] Issues related to food preparation, sharing, distribution, indigenous knowledge, attitudes and practices should be considered.
## Multisectoral partnership/coordination
Prevention and/or management of overweight/obesity in children is a shared responsibility between different sectors. No single sector will address this complex, multifaceted problem. African member states should be willing to take responsibilities through preparation and enhancement of policies across all sectors.The health sector needs to initiate, implement, and innovate primary and secondary overweight/obesity preventive measures. The education sector needs to integrate and implement nutrition-related courses in primary school curriculumwith content delivery predicated on the building of teachers' capacities to deliver nutrition education to children. The agriculture sector needs to emphasize on the production of local nutrient-dense foods at an affordable cost and the periodic revision of import trade policies.Food regulatory bodies are tasked to set and enforce import requirements and regulations to ensure that imported foods meet agreed quality standards.In addition, food and beverage industry need to focus on healthy product development, by reformulating nutrient-dense food categories to deliver better diets for all. Due to industrial development and economic changes many people have shifted their food habits from whole cereals, legumes, fruits and vegetables towards purchasing and consuming more processed foods and drinks. [bib_ref] Study of nutrition habits in primary school students, Alejandra [/bib_ref] Therefore, this sector needs a high level of commitment in manufacturing healthy products that are affordable and available to all groups of consumers.
## Social behavior change communication
One of the most promising strategies to prevent childhood obesity is social behavior change communications. This strategy includes individual counseling, mass media campaigns, and education sessions. If these are targeted to relevant audiences, like school children, schools, nongovernmental organizations, and decision-makers, the expected outcomes may be achieved. 32 An example of this approach is reflected in Tanzania's national nutrition social behavior change communication (SBCC) strategy which aims at raising awareness through an increase in knowledge, attitude, and skill training. It also aims at creating quality nutrition services that will favor the demand of consumers, other beneficiaries and increase access to quality communication materials, SBCC guidelines, protocols, and other tools to the district level.SBCC needs to be developed, pre-tested, and disseminated to the target population. A review of literature in developing countries showed success in integrating social behavior change communication interventions with nutritional specific programs. [bib_ref] Impact of social and behavior change communication in nutrition specific interventions on..., Kennedy [/bib_ref] For example, a study conducted in Malawi used SBCC strategy in a supplementary feeding program and showed promising effect. [bib_ref] Effective delivery of social and behavior change communication through a Care Group..., Wilner [/bib_ref]
# Conclusion
This review found an increased and varied prevalence of overweight and obesity in school children up to above 40% across African countries in ten years period. However, the studies focusing on their linkages with home and school environment are scarce, therefore, this calls for more research attention and informed policy change in this area. High socio-economic status, female gender, urban residence and children in private schools associated significantly with increased prevalence of overweight/obesity in children. The presence of high energy-dense foods in school and home environments contribute to unhealthy dietary choices in school children. Most of the articles reviewed were methodologically limited as they involved "one-off" crosssectional design studies as opposed to longitudinal, case-control, and controlled randomized trials. However, some studies on food environment used qualitative approach, which is strong in highlighting the subjects' perceptions, ideas, and opinions. This review was restricted to school and home physical food environment, (availability and accessibility). Other components were not considered. It focused on a topic which, in most cases, receives low attention in nutrition research. This review was limited to 10 years period, and used English, a language conversant to authors and to the majority of readers.
## Abbreviations
# Author contributions
All authors made a significant contribution to the reported work, from the conception, study design, execution, acquisition of data and interpretation. In all these areas; they took part in drafting, revising and critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
# Funding
No funding was required to prepare this manuscript.
# Disclosure
The authors declare that they have no conflicts of interest.
[fig] Figure 1: Flow chart of the selection process for inclusion of eligible studies. Pacific et al Dovepress submit your manuscript | www.dovepress.com DovePress Pediatric Health, Medicine and Therapeutics 2020:11 [/fig]
[table] Table 1: Summary of Evidence on Prevalence of Overweight and Obesity in School Children in Selected African Countries [/table]
[table] Table 2: Summary of Studies on Influence of School and Home Environment on Children Food Choice in Selected African Countries Qualitative Parental control dictates what a child eats; eating out exposes a child to junk foods; peer pressure influence on food choice; energy-dense foods are readily available in school shops. Food vendors around school environments formed the main food suppliers, followed by school shops, canteens. Main foods supplied were snacks and soft drinks, while fruits, vegetables and dairy products were less available. Home Qualitative Role of caregivers in preparing healthy foods through the availability of home gardens; adolescent perception on healthy foods; barriers to access healthy foods. The presence of healthy food varieties in the school tuck shop influenced a positive attitude and perceptions towards healthy eating although unhealthy options should also be available. Abrahams et al, 2011 67 South Africa School Quasi-experiment Children who brought lunch boxes to school comprised of healthy food options had lower BMI compared to children who bought food items from school shops. Kupolati et al, 2017 20 South Africa School Qualitative Teachers perceived that nutrition education can influence positive healthy eating habits in the school environment. provides less options of health foods like fruits and vegetables but more of unhealthy food items like chips, sweets, chocolates and biscuits which are preferred by children. Confectionaries and deep-fried foods, soft drinks and dessertswere most commonly sold (by more than 75%) followed by main meals from canteens and food vendors; healthy foods like dairy products and fruits were also available Abbreviation: BMI, body mass index. [/table]
|
Is the gene encoding Chibby implicated as a tumour suppressor in colorectal cancer ?
Background:A novel member of the Wnt signalling pathway, Chibby, was recently identified. This protein inhibits Wnt/β-catenin mediated transcriptional activation by competing with Lef-1 (the transcription factor and target of β-catenin) to bind to β-catenin. This suggests that Chibby could be a tumour suppressor protein. The C22orf2 gene coding Chibby is located on chromosome 22, a region recurrently lost in colorectal cancer. Activation of the Wnt pathway is a major feature of colorectal cancer and occurs through inactivation of APC or activation of β-catenin. All of this led us to analyse the possible implication of Chibby in colorectal carcinogenesis.Methods: First, 36 tumour and matched normal colonic mucosa DNA were genotyped with five microsatellite markers located on chromosome 22 to search for loss of heterozygosity. Then, mutation screening of the C22orf2 coding sequence and splice sites was performed in the 36 tumour DNA. Finally, expression of Chibby was analysed by quantitative RT-PCR on 10 patients, 4 with loss of heterozygosity (LOH) on chromosome 22.Results:Loss of heterozygosity involving the C22orf2 region was detected in 11 out of 36 patients (30%). Sequencing analysis revealed a known variant, rs3747174, in exon 5: T321C leading to a silent amino acid polymorphism A107A. Allelic frequencies were 0.69 and 0.31 for T and C variants respectively. No other mutation was detected. Among the 10 patients studied, expression analysis revealed that Chibby is overexpressed in 2 tumours and underexpressed in 1. No correlations were found with 22q LOH status.Conclusion: As no somatic mutation was detected in C22orf2 in 36 colorectal tumour DNA, our results do not support the implication of Chibby as a tumour suppressor in colorectal carcinogenesis. This was supported by the absence of underexpression of Chibby among the tumour samples with 22q LOH. The implication of other Wnt pathway members remains to be identified to explain the part of colorectal tumours without mutation in APC and β-catenin.
# Background
Identifying components of the Wnt signalling pathway has been at the forefront of cancer biology since a link was made between Wnt, the mammalian homologue of the fruitfly Wingless (Wg), and the development of cancer. Acting through a core set of proteins that are highly conserved in evolution, this pathway regulates the ability of the oncoprotein β-catenin to activate transcription of specific target genes. This regulation, in turn, results in changes in expression of genes that modulate cell fate, proliferation and apoptosis [bib_ref] The promise and perils of Wnt signaling through beta-catenin, Moon [/bib_ref]. Recently, Takemaru et al. identified a novel human protein, named Chibby, that interacts with the carboxy-terminal transcription activation domain of β-catenin [bib_ref] Chibby, a nuclear beta-catenin-associated antagonist of the Wnt/Wingless pathway, Takemaru [/bib_ref]. Chibby is a nuclear protein of 126 amino acids with coiled-coil domains and is conserved from Drosophila to Human. It has been shown that Chibby antagonizes the Wnt signalling pathway by inhibition of the transcription protein complex comprising β-catenin. This result suggests that Chibby could act as a tumour suppressor protein.
In colorectal cancer, the activation of the Wnt signalling pathway occurs in more than 60% of tumours through the inactivation of the APC tumour suppressor gene by mutations and allelic losses, or through the presence of β-catenin activating mutations [bib_ref] Sequence of molecular genetic events in colorectal tumorigenesis, Laurent-Puig [/bib_ref]. The C22orf2 gene encoding Chibby is located on 22q13.1. This chromosome region is frequently lost in colorectal cancer suggesting the existence of a tumour suppressor gene that remains to be identified.
The putative function and the location of the C22orf2 gene led us to analyse the possible implication of C22orf2 as a tumour suppressor gene in colorectal carcinogenesis. First, the allelic status of chromosome 22 was established on 36 colorectal tumour and matched normal colonic mucosa DNA, second, mutation analysis of the C22orf2 gene was performed on tumour DNA, and third, expression analysis of Chibby was studied in few patients.
# Methods
## Patients, sample collection and nucleic acid extraction
Tumour samples and matched normal colonic mucosa were collected from 36 patients (23 women, 13 men) hospitalised in the surgical department of Laennec Hospital in Paris between 1997 and 1999. The mean age of patients was 69.9 years old, range [56. ]. An histological HES staining was performed before DNA extraction and only tumour fragments with more than 70% of tumour cells were retained. Tumours were all classified as adenocarcinoma. Twenty-nine tumours were classified as well differentiated, and 7 as poor differentiated. No tumour samples showed a microsatellite instability phenotype. Tumours were located in proximal colon in 9 cases, in distal colon in 21 cases and 6 were located in rectum. Accord-ing to TNM classification, tumours were classified in stage I in 1 case, stage II in 17, stage III in 8 and stage IV in 10 cases. Samples were immediately frozen in liquid nitrogen and stored at -80°C. Informed consent was signed according to French laws. DNA extraction was performed using the QIAamp ® DNA Mini Kit (Qiagen, Courtaboeuf, France) for all patients in 1999, and stored at -20°C. Tumour samples and matched normal colonic mucosa were available for RNA extraction for 10 patients. RNA isolation was performed recently with RNeasy ® Mini Kit (Qiagen) and stored at -80°C. Quality of RNA was determined by electrophoresis through agarose gel stained with ethidium bromide. Intensity of 18S and 28S RNA bands was estimated under UV light. Measuring UV absorbance at 260 nm was performed to quantify RNA.
## Genotyping
Thirty-six tumour and matched normal colonic mucosa DNA were genotyped with five microsatellite markers located on chromosome 22. Amplification of the five markers was performed by multiplex PCR using 50 ng DNA in a final reaction volume of 12.5 µL using 6.25 µL of 2X Qiagen Multiplex PCR Master Mix (Qiagen) and variable concentration of primers as given in [fig_ref] Table 1: Microsatellite markers analysed on chromosome 22 [/fig_ref] (Goasguen et al, unpublished data). Primers are from Linkage Mapping Set (Applied Biosystems, Courtaboeuf, France) and labelled on forward primer by VIC, NED or 6-FAM. Amplified fragments were analysed after dilution on an ABI 310 Genetic Analyzer (Applied Biosystems). Allelic imbalance (AI) was analysed by Genotyper 2.1 ® software (Applied Biosystems) with an automatic data-processing trace program described previously. AI was defined for each tumor as α = (NL/NS)/(TL/TS) where L is the intensity of the larger allele and S the intensity of the smaller allele in normal DNA (N) or tumour DNA (T). When α(loss of the small allele) or 1/α (loss of the large allele) was ≤ 0.5, LOH was ascertained.
# Sequencing analysis
PCR amplification of the four C22orf2 coding exons and intron-exon junctions was performed as described in [fig_ref] Table 2: PCR conditions for sequencing analysis of the C22orf2 gene [/fig_ref]. Due to the small size of intron 3, exons 3 and 4 were amplified together. 100 ng of tumour DNA were used in a final reaction volume of 30 µL. For exons 2 and 3-4, 0.75 U of AmpliTaq ® polymerase (Applied Biosystems) were used with 3 µL of 10X corresponding buffer, 1.5 mM MgCl2, 800 µM of dNTP mix and 0.3 µM of each forward and reverse primers. Program was as follows : initial step at 94°C 5 min, 35 cycles of 94°C 30 s, 58°C 30 s and 72°C 30 s, and final step 72°C 7 min. For exon 5, 0.75 U of HotStarTaq ® polymerase (Qiagen) were used with the same protocol except the initial step at 95°C 10 min and the annealing temperature at 56°C. PCR products were purified by centrifugation on a Sephadex G-100 ® (Amersham Biosciences, Orsay, France) filter loaded in a MultiScreen ® plate (Millipore, Molsheim, France). Then direct sequencing was performed in both forward and reverse orientation from 2 µL of purified products using 2 µL of 2.5X BigDye™ Terminator v3.1 Ready Reaction Cycle Kit (Applied Biosystems) with 3 µL of 5X corresponding buffer and 0.5 µM of primer in a final reaction volume of 20 µL. Products were purified on Sephadex G-50 ® (Amersham Biosciences) and analysed on an ABI 310 Genetic Analyzer (Applied Biosystems). Sequence data were aligned using AutoAssembler™ 2.0 software (Applied Biosystems).
## Quantitative rt-pcr
Three µg of RNA were reverse transcribed in a final volume of 50 µL using the High Capacity cDNA Archive Kit with random primers (Applied Biosystems). Reverse transcribed samples were diluted 20 fold in water and stored at -20°C. For each sample, 5 µL of cDNA, corresponding to 15 ng of reverse transcribed RNA, were analysed by SYBR Green PCR, in triplicate, using the ABI PRISM ® 7900HT Sequence Detection System (Applied Biosystems). Q-PCR was performed in a final volume of 12 µL comprising 1X Master Mix SYBR Green (ABgene, Courtaboeuf, France) and 0.3 µM of forward and reverse primers for Chibby (Fwd: 5'-GGAGAAAACCAAGATTCCAG-3' ; Rev: 5'-CCAACACAACCCAACAGAG-3'). Levels of RNA expression were determined using the SDS software version 2.1 (Applied Biosystems) according to the 2 -∆∆Ct method. Briefly, expression results of a gene were normalised to internal control ribosomal 18S and relatively to a calibrator, consisting in the mean expression level of the corresponding gene in normal colonic mucosa samples as follows : 2 -∆∆Ct = 2 -((Ct Chibby -Ct 18S)tumour sample) -(Ct Chibby -Ct 18S) calibrator)) . Results are given in [fig_ref] Table 3: Expression analysis of Chibby by Quantitative RT-PCRPatient 2 -∆∆Ct Expression of Chibby... [/fig_ref] , they express the n-fold ratio of the gene expression in a sample compared to the mean of normal tissues. A ratio ≤ 0.66 has been considered as underexpression of Chibby and ≥ 1.5 as overexpression.
# Results
## Loss of heterozygosity analysis on chromosome 22
Five dinucleotide microsatellite markers spread over the entire chromosome 22 were genotyped in order to determine the existence of a loss of heterozygosity (LOH). The human gene coding Chibby, C22orf2, is localised between the markers D22S283 and D22S423 [fig_ref] Table 1: Microsatellite markers analysed on chromosome 22 [/fig_ref]. Thirty-six tumour and matched normal colonic mucosa DNA were amplified by multiplex PCR. A deletion of the entire chromosome 22 arm was observed in 9 cases, a distal deletion including C22orf2 was observed in 2 cases [fig_ref] Figure 1: Microsatellite markers used for genotyping of chromosome 22 [/fig_ref] and an interstitial deletion encompassing the D22S315 was observed in one case. Thus, among the 36 tumours, 11 (30%) demonstrated LOH involving the C22orf2 region.
## Sequencing analysis of c22orf2
The C22orf2 gene spreads over 17.2 kb and comprises five exons. ORF starts at exon 2 leading to a 381 bp cDNA. Coding sequence and intron exon junctions were analysed by direct sequencing on the 36 tumour DNA to search for sequence variations. A known variant, rs3747174, was observed in exon 5 leading to a T321C transition, corresponding to a silent polymorphism A107A. Allelic frequencies were 0.69 and 0.31 for T and C variants respectively. The distribution of the different genotypes is in agreement with Hardy and Weinberg equilibrium. No other somatic mutation was detected either in coding exons or in splice sites.
## Expression analysis of chibby
The difference of the level of expression of Chibby between tumour samples and normal colonic mucosa samples was analysed by quantitative RT-PCR, using ribosomal 18S as internal control. Both samples were available for 10 patients, 4 of which presented LOH on chromosome 22. Q-PCR expression analysis revealed that Chibby is overexpressed in 2 tumours and underexpressed in 1 [fig_ref] Table 3: Expression analysis of Chibby by Quantitative RT-PCRPatient 2 -∆∆Ct Expression of Chibby... [/fig_ref]. No correlations were found with 22q LOH status.
# Discussion
The aim of this study was to analyse the possible implication of Chibby in colorectal carcinogenesis. Genotyping analysis was performed on chromosome 22 to search for loss of heterozygosity. Among the 36 patients, 12 (33%) showed LOH including the C22orf2 region. Mutation analysis was performed on C22orf2 coding sequence and splice sites: no somatic mutation was detected. However, a known variant, rs3747174, was observed in exon 5: T321C, corresponding to A107A. In the dbSNP database http://www.ncbi.nlm.nih.gov/SNP/, this variant was detected with a set of 52 chromosomes and allelic frequencies were estimated from 1496 chromosomes at 0.63 and 0.37 respectively which is similar to that observed in this present series, i.e. 0.67 and 0.31 respectively [bib_ref] Gene-based SNP discovery as part of the Japanese Millennium Genome Project: identification..., Haga [/bib_ref]. Furthermore, quantitative RT-PCR expression analysis showed that Chibby is overexpressed in 2 tumours, 1 of which showing LOH at C22orf2 locus, and Chibby is underexpressed in 1 tumour showing no 22q LOH. Taking together, these results do not support a putative epige-netic modification, i.e. methylation of the C22orf2 promoter, that could repress gene expression as another mechanism of gene inactivation than mutation. Thus, Chibby does not seem implicated as a tumour suppressor in colorectal carcinogenesis.
# Conclusions
The APC gene was found mutated in several series of colorectal tumours with a frequency of 60%. Furthermore, in a series of tumours lacking APC mutations, 48% presented a mutation in the β-catenin regulatory domain [bib_ref] Sequence of molecular genetic events in colorectal tumorigenesis, Laurent-Puig [/bib_ref]. Thus, more than one gene could be implicated in colorectal carcinogenesis through the activation of the Wnt signalling pathway. The recently identified function of Chibby in this pathway supports the idea that it could act as a tumour suppressor [bib_ref] Chibby, a nuclear beta-catenin-associated antagonist of the Wnt/Wingless pathway, Takemaru [/bib_ref]. However, we did not detect mutation in a series of 36 colorectal tumours, suggesting that the role of Chibby in colorectal carcinogenesis is probably weak. Other genes remain to be studied to explain the part of colorectal tumours without mutation in APC and β-catenin.
## Competing interests
None declared.
# Authors' contributions
SG carried out genotyping and expression analysis, trained DT and drafted the manuscript. DT carried out the sequencing analysis. AL and NG have developed multiplex PCR with microsatellite markers. AB provided patient samples. PB and PLP conceived of the study, participated in its design and coordination, and drafted the manuscript. All authors read and approved the final manuscript. Microsatellite markers used for genotyping of chromosome 22
[fig] Figure 1: Microsatellite markers used for genotyping of chromosome 22. Genotyper ® graphs obtained after multiplex PCR of the five microsatellite markers on chromosome 22 are shown on A, B, C, D, E. In each case the upper graph corresponds to the tumour DNA and the lower graph to the matched normal colonic mucosa DNA. Peaks are labelled with: name of markers, size of PCR products (bp) and peak intensity. A and B show no allelic loss for D22S420 and D22S315, whereas C, D and E show allelic loss for DS22S283, D22S423, D22S274. These results suggest a distal deletion of chromosome 22 including the C22orf2 gene. [/fig]
[table] Table 1: Microsatellite markers analysed on chromosome 22 [/table]
[table] Table 2: PCR conditions for sequencing analysis of the C22orf2 gene [/table]
[table] Table 3: Expression analysis of Chibby by Quantitative RT-PCRPatient 2 -∆∆Ct Expression of Chibby in tumour tissues compared to normal tissues LOH on chromosome 22 [/table]
|
The role of autophagy and ferroptosis in sensorineural hearing loss
CITATIONSun Y, Zou S, He Z and Chen X (2022) The role of autophagy and ferroptosis in sensorineural hearing loss.Hearing loss has become a common sensory defect in humans. Because of the limited regenerative ability of mammalian cochlear hair cells (HCs), HC damage (caused by ototoxic drugs, aging, and noise) is the main risk factor of hearing loss. However, how HCs can be protected from these risk factors remains to be investigated. Autophagy is a process by which damaged cytoplasmic components are sequestered into lysosomes for degradation. Ferroptosis is a novel form of non-apoptotic regulated cell death involving intracellular iron overloading and iron-dependent lipid peroxide accumulation. Recent studies have confirmed that autophagy is associated with ferroptosis, and their crosstalk may be the potential therapeutic target for hearing loss. In this review, we provide an overview of the mechanisms of ferroptosis and autophagy as well as their relationship with HC damage, which may provide insights for a new future in the protection of HCs. KEYWORDS autophagy, ferroptosis, ferritinophagy, sensorineural hearing loss, hair cell Frontiers in Neuroscience 01 frontiersin.org
# Introduction
As a common sensory disorder in humans, hearing loss affects 1.5 billion people around the world . Sensorineural hearing loss (SNHL), an important type of hearing loss, refers to pathological changes in the cochlear hair cells (HCs), auditory nerves, or central auditory system. Ototoxic drugs, aging, and noise are common pathogenic factors of hearing loss that can cause damage or loss of HCs [bib_ref] Review of environmental factors affecting hearing, Mills [/bib_ref]. Unfortunately, adult mammalian cochlear HCs are not capable of regeneration; consequently, irreversible damage to HCs can result in permanent deafness [bib_ref] A human induced pluripotent stem cell-based modular platform to challenge sensorineural hearing..., Zine [/bib_ref]. Protecting HCs is, therefore, a key focus of auditory research. However, the detailed mechanism underlying cochlear HC damage remains largely unknown.
In recent years, autophagy and ferroptosis have attracted much attention, and the modulation of autophagy and ferroptosis is a very attractive potential approach for the future treatment of SNHL. Autophagy is a highly conserved degradation system in eukaryotic cells that can remove damaged organelles and macromolecular proteins and be induced by reactive oxygen species (ROS) accumulation [bib_ref] Autophagy in stem cells: repair, remodelling and metabolic reprogramming, Boya [/bib_ref]. In the auditory system, especially in HCs, previous studies reported the relationship between autophagy and hearing loss [bib_ref] Autophagy regulates the survival of hair cells and spiral ganglion neurons in..., Guo [/bib_ref] [bib_ref] Autophagy impairment as a key feature for acetaminophen-induced ototoxicity, Zhao [/bib_ref] , specifically, that moderately activating autophagy can reduce HC damage in both cochlear HCs and HEI-OC1 cells (a mouse auditory cell line) after neomycin or gentamicin injury [bib_ref] Autophagy protects auditory hair cells against neomycin-induced damage, He [/bib_ref]. Ferroptosis is a non-apoptotic mechanism of cell death that mainly relies on the ironmediated production of ROS and follows plasma membrane damage [bib_ref] The role of iron and reactive oxygen species in cell death, Dixon [/bib_ref]. Generally, ferroptosis accompanies lipid peroxidase, impaired antioxidant capacity, and increased intracellular iron accumulation . New research has found that the development and progression of many diseases are related to ferroptosis [bib_ref] Emerging roles of ferroptosis in cardiovascular diseases, Wang [/bib_ref]. Recently, several studies have revealed the link between ferroptosis and hearing loss and proved that the inhibition of ferroptosis may alleviate HC damage [bib_ref] Autophagydependent ferroptosis contributes to cisplatin-induced hearing loss, Jian [/bib_ref].
In this paper, we review recent studies about the role of autophagy and ferroptosis in HC damage. We also focus on the molecular regulatory mechanism of autophagy and ferroptosis in vitro and in vivo. Additionally, in this review, we discuss the regulatory impact of autophagy on ferroptosis and further discuss whether regulating ferroptosis through autophagy could be a potential measure for the protection of HCs.
## The role of autophagy in hair cell damage
Autophagy is not only a physiological process of degradation for long-lived proteins and damaged organelles in eukaryotic cells but also an important adaptive mechanism in the response to cellular stress (e.g., infection, poisoning, and hypoxia) [bib_ref] The roles of reactive oxygen species (ROS) and autophagy in the survival..., Chen [/bib_ref]. Under normal physiological conditions, autophagy is essential for the maintenance of cell homeostasis, while under some stress conditions, autophagy will be activated and promoted. Autophagy is divided into macroautophagy, microautophagy, and chaperonemediated autophagy [bib_ref] Autophagy: a Druggable Process, Morel [/bib_ref]. Generally speaking, macroautophagy, as the main process for cells to remove damaged organelles and other related fragments, is referred to as "autophagy" [bib_ref] The machinery of macroautophagy, Feng [/bib_ref]. Autophagy takes place throughout the dynamic development of cells and plays the function of a "double-edged sword" [bib_ref] A double-edged sword with therapeutic potential: an updated role of autophagy in..., Wei [/bib_ref]. In other words, while autophagy confers a pro-survival effect, allowing cells to fight against cellular stressors such as excess ROS [bib_ref] The roles of reactive oxygen species (ROS) and autophagy in the survival..., Chen [/bib_ref] , under certain circumstances, excessive autophagy can also cause autophagic cell death [bib_ref] Autophagic cell death exists, Clarke [/bib_ref].
## The role of autophagy in age-related hearing loss
Age-related hearing loss (ARHL) is the most common form of aging in the auditory system and clinically manifests as progressive bilateral symmetrical hearing loss [bib_ref] Age-related hearing loss. Cold Spring Harb, Bowl [/bib_ref]. As a major cause of sensory disturbances, ARHL limits social interaction among older individuals and thus leads to cognitive function decline, depression, anxiety, and loneliness [bib_ref] Age-related hearing loss and cognitive decline: MRI and cellular evidence, Jafari [/bib_ref]. Oxidative stress theory is the most common hypothesis regarding aging [bib_ref] Genetic analysis of ageing: role of oxidative damage and environmental stresses, Martin [/bib_ref]. With increasing age, the production of ROS increases, while body antioxidant capacity gradually decreases. The level of autophagy decreases with age [bib_ref] Age-Related Hearing Loss in C57BL/6J Mice Is Associated with Mitophagy Impairment in..., Youn [/bib_ref] , and moderate upregulation of autophagy can promote the survival of aging HCs by scavenging damaged mitochondria, which may suggest that autophagy impairment is a key factor in ARHL [bib_ref] Nrf2 activation induces mitophagy and reverses Parkin/Pink1 knock down-mediated neuronal and muscle..., Gumeni [/bib_ref]. Moderately promoting autophagic activity may, therefore, advance the viability of aging HCs.
Recently, some molecules have been reported to regulate the level of autophagy in ARHL. MicroRNAs (miRNAs) are a kind of short, endogenous, and highly evolutionarily conserved single-stranded RNAs. miRNAs can regulate the expression of target genes by inducing the degradation and inhibiting the translation of target mRNA. miR-34a blocks autophagy flux by inhibiting the expression of the autophagy-related gene 9A (ATG9A) and promotes the death of senescent HEI-OC1 cells [bib_ref] Activation of miR-34a impairs autophagic flux and promotes cochlear cell death via..., Pang [/bib_ref]. Additionally, miR-34a mediates autophagy through more than one target. In HEI-OC1 cells, siRNA was used to downregulate the expression of Sirtuin 1 (SIRT1), which can impair autophagy by significantly decreasing the conversion of microtubule-associated protein 1 light chain 3 (LC3) I to LC3 II and influencing the deacetylation of ATG9A, inducing the accumulation of p62 and HEI-OC1 cell death . One study suggested that miR-34a targeted SIRT1, protected cochlear HCs, and delayed ARHL by regulating mitochondria-specific autophagy (mitophagy) . In addition, Forkhead box G1(FOXG1) protects HCs from oxidative damage by activating autophagy during cell senescence .
As a selective kind of autophagy, mitophagy plays an important role in ARHL pathological changes. Mitophagy refers to the process of selectively degrading damaged mitochondria and maintaining mitochondrial homeostasis. Since mitochondria are the "energy factories" and free radical metabolic centers in cells, dysfunction caused by mitochondrial damage leads to metabolic disorders of ROS. Excessive ROS can cause oxidative damage to mitochondrial lipids, DNA, and proteins, making mitochondria produce more ROS and eventually triggering apoptosis [bib_ref] The pathways of mitophagy for quality control and clearance of mitochondria, Ashrafi [/bib_ref]. However, to prevent cellular damage, mitophagy will be activated to preserve a population of healthy mitochondria. Researchers found that mitophagy is involved in aging HCs.
Prohibitin 2 (PHB2), a protein receptor on the intima of mitochondria, is exposed when the mitochondrial outer membrane is ruptured and then interacts with the autophagy protein LC3 II to activate mitophagy. [bib_ref] The expression of PHB2 in the cochlea: Possible relation to age-related hearing..., Yu [/bib_ref] found that PHB2 expression was increased along with mitophagy activation in mice with ARHL, indicating that PHB2 may be involved in mitophagy. However, the exact mechanism underlying the role of PHB2 in ARHL is unclear and thus needs to be explored further . BCL2 interacting protein 3 like/Nip3like protein X (BNIP3L/NIX) is a protein located in the outer mitochondrial membrane that participates in mitophagy by promoting the formation of autophagosomes to engulf target mitochondria . BNIP3L/NIX was significantly downregulated in both in vivo and in vitro models of aging, while overexpression of BNIP3L/NIX alleviated HC senescence by promoting mitophagy [bib_ref] BCL2 Interacting Protein 3-like/NIX-mediated mitophagy plays an important role in the process..., Kim [/bib_ref].
## The role of autophagy in drug-induced hearing loss
Chemotherapeutic agents (e.g., cisplatin) and aminoglycosides (e.g., neomycin and gentamicin) are two major classes of ototoxic drugs. These drugs can cause bilateral and irreversible SNHL. Because of their side effects, the application of ototoxic drugs is limited, and hearing loss prevention following ototoxic drug treatment has long attracted researchers' attention. ROS have been identified as one of the classical causes of SNHL mediated by cisplatin and aminoglycosides [bib_ref] A reduced form of nicotinamide riboside protects the cochlea against aminoglycoside-induced ototoxicity..., Fang [/bib_ref] and can be generated after ototoxic drug treatment [bib_ref] Wnt activation protects against neomycin-induced hair cell damage in the mouse cochlea, Liu [/bib_ref] [bib_ref] Meclofenamic acid reduces reactive oxygen species accumulation and apoptosis, inhibits excessive autophagy,..., Li [/bib_ref]. If ROS are generated in excess, they will damage the antioxidant defense capacity of HCs and induce cell apoptosis.
Autophagy is a common cellular response when cells are exposed to stress stimuli such as ototoxic drugs. Previous studies showed that ROS can induce autophagy in auditory cells after cisplatin and aminoglycoside injury [bib_ref] Reactive oxygen species in chick hair cells after gentamicin exposure in vitro, Hirose [/bib_ref] [bib_ref] PRDX1 activates autophagy via the PTEN-AKT signaling pathway to protect against cisplatin-induced..., Liu [/bib_ref]. Autophagic activity increases significantly in HCs following treatment with neomycin or cisplatin [bib_ref] Loss of STAT1 protects hair cells from ototoxicity through modulation of STAT3,..., Levano [/bib_ref] [bib_ref] Autophagy protects auditory hair cells against neomycin-induced damage, He [/bib_ref]. The activation of autophagy has been observed to promote HC survival after neomycin treatment, and blocking the activation of autophagy can increase ROS levels and induce HC apoptosis [bib_ref] Autophagy protects auditory hair cells against neomycin-induced damage, He [/bib_ref]. Therefore, exploring the effect of autophagy on HC protection along with the underlying mechanism may have therapeutic implications for the prevention and treatment of ototoxic drug-induced hearing loss.
Autophagy is regulated by many factors during ototoxic drug damage. The overexpression of YTHDF1 (the YT521-B homology N6-methyladenosine RNA binding protein) could enhance the activation of autophagy by promoting the translation of the autophagy-related gene ATG14 (autophagy-related gene 14), while a deficiency thereof suppressed autophagy and increased the loss of HCs after cisplatin damage . Overexpression of TFEB (transcription factor EB), a transcription factor, activated autophagy in cochlear HCs and HEI-OC1 cells and subsequently alleviated cisplatin-induced apoptosis [bib_ref] Trehalose protects against cisplatin-induced cochlear hair cell damage by activating TFEBmediated autophagy, Li [/bib_ref]. Upregulation of autophagy by phosphorylation of the Ser9 site of GSK-3β(glycogen synthase kinase-3β), a multifunctional serine/threonine protein kinase, could attenuate cisplatin-induced ototoxicity, which may be related to the activation of the WNT/β-catenin signaling pathway . PINK1 (phosphatase and tensin homolog induced putative kinase 1)/Parkin-mediated mitophagy can inhibit mitochondrial proteotoxicity and maintain mitochondrial protein homeostasis. Research reported that in HEI-OC1 cells, neomycin attenuated PINK1/Parkin-mediated mitophagy by promoting ATF3 (activating transcription factor 3) expression, and damaged cochlear HCs could be partially rescued when mitophagy was induced . After cisplatin treatment, ROS aggravates the damage and promotes the activation of PINK1/Parkin-related mitophagy to clear dysfunctional mitochondria and inhibit the activation of JNK (c-Jun N-terminal kinase)-related apoptotic pathways to protect HEI-OC1 cells against cisplatin-induced damage .
## The role of autophagy in noise-induced hearing loss
Noise is one of the main causes of SNHL. Following noise exposure, the accumulation of ROS contributing to the pathogenesis of noise-induced loss of sensory HCs is well accepted. Noise-induced stress increases the level of ROS in outer HCs and then leads to HC damage, which is the key to metabolic injury [bib_ref] Noise-induced loss of hair cells and cochlear synaptopathy are mediated by the..., Hill [/bib_ref]. In addition, ROS induce inflammation by producing cytokines and damage cochlear HCs [bib_ref] Blockade of interleukin-6 signaling suppressed cochlear inflammatory response and improved hearing impairment..., Wakabayashi [/bib_ref]. However, ROS can also induce autophagy to perform cellular defense functions [bib_ref] Eat-me: autophagy, phagocytosis, and reactive oxygen species signaling, Vernon [/bib_ref].
Following noise exposure, autophagy is triggered to limit pathological changes to HCs. [bib_ref] Plasma metabolomic profiling in workers with noise-induced hearing loss: a pilot study, Miao [/bib_ref] found that two autophagy metabolites, phosphatidylethanolamine (PE) and phosphatidylcholine, were significantly decreased in the plasma samples of patients suffering from noiseinduced hearing loss (NIHL). They confirmed that PI3K (phosphatidylinositide 3-kinase), AKT (protein kinase B), and ATG5 (autophagy-related gene 5) were significantly downregulated in NIHL patients by using real-time quantitative PCR .found that increasing autophagy activity by using rapamycin can inhibit ROS accumulation and reduce noise-induced HC loss. In contrast, treatment with either the autophagy inhibitor 3-methyladenine or LC3 II siRNA can exacerbate noise-induced HC loss and NIHL . Recently, pejvakin has been reported to regulate autophagy in NIHL. Pejvakin is a peroxisome-associated protein from the gasdermin family that helps to maintain the normal three-dimensional ciliated ladder structure and mechanical transduction of HCs [bib_ref] Pejvakin, a candidate stereociliary rootlet protein, regulates hair cell function in a..., Kazmierczak [/bib_ref]. In response to sound overstimulation, pejvakin promotes pexophagy (the autophagic degradation of peroxisomes) by recruiting LC3 II directly and protects the auditory HCs against oxidative stress . Based on the above studies, the importance of autophagy in NIHL is worthy of further exploration.
The role of ferroptosis in hair cell damage [bib_ref] Ferroptosis: an iron-dependent form of nonapoptotic cell death, Dixon [/bib_ref] identified a novel form of regulated cell death, denoted "ferroptosis." The morphology and biochemistry of ferroptosis differ from other forms of cell death. Morphologically, mitochondrial cristae loss and outer membrane rupture are characteristic signatures of ferroptotic cells . Biochemically, ferroptosis is characterized by its association with the accumulation of iron and the subsequent production of ROS, as well as lipid peroxidation . Once the ability of the antioxidant system to scavenge overexpressed ROS is exceeded, oxidative stress will occur, resulting in lipid peroxidation, which damages cellular structures and leads to cell death. The labile iron pool, moreover, can facilitate a Fenton reaction and peroxidize polyunsaturated fatty acids (PUFAs) to generate lipid peroxides, resulting in plasma membrane rupture and eventually, cell death [bib_ref] Ferroptosis turns 10: emerging mechanisms, physiological functions, and therapeutic applications, Stockwell [/bib_ref]. In brief, iron-dependent ROS production results in lipid peroxidation and ultimately leads to membrane damage, which is the core mechanism of ferroptosis. ROS produced by the Fenton reaction, PUFAs, and lipid peroxidation can promote ferroptosis, but the system Xc − /glutathione(GSH)/glutathione peroxidase 4(GPX4) axis can inhibit ferroptosis by influencing the redox status of cells [bib_ref] Ferroptosis at the intersection of lipid metabolism and cellular signaling, Liang [/bib_ref].
Since ferroptosis was discovered, some evidence has shown the close relationship between ferroptosis and many pathologies. For example, ferroptosis is widely believed to function as a tumor-suppression mechanism. Phosphoglycerol dehydrogenase, by binding and regulating PCBP2 [Poly(rC) binding protein 2] protein expression to promote the mRNA stability of SLC7A11 (the catalytic subunit of system Xc − ), can inhibit cell ferroptosis and ultimately promote the malignant progression of bladder cancer. The ferroptosis inducer dihydroartemisinin intensively strengthens the cytotoxicity of cisplatin for pancreatic ductal adenocarcinoma cells, which may act as a promising therapeutic strategy for drug-resistant tumors [bib_ref] DHA exhibits synergistic therapeutic efficacy with cisplatin to induce ferroptosis in pancreatic..., Du [/bib_ref].
Similarly, ferroptosis also plays an important role in HC injury. Neomycin induced intracellular Fe 2+ increases and lipid peroxidation in the HEI-OC1 cell line and cochlear HCs. After cisplatin treatment, the mitochondria in HEI-OC1 cells showed mitochondrial shrinkage and loss of mitochondrial crests-typical characteristics of ferroptotic cells . Moreover, ferroptosis may be a cause of auditory cortex degeneration, leading to central presbycusis. Levels of acyl-CoA synthetase longchain family member 4 (ACSL4) and transferrin receptor 1 were increased in the auditory cortex of rat models that mimic aging induced by D-galactose, and removing excess iron from cells by deferoxamine treatment inhibited ferroptosis and delayed cellular senescence . However, the role of ferroptosis in aging HCs has not been reported.
As mentioned above, ferroptosis occurs when lipid ROS production exceeds the antioxidant capability. Glutathione peroxidase 4 (GPX4) can convert lipid hydroperoxides to lipid alcohols by using reduced GSH as a cofactor, reducing ROSinduced damage to cells [bib_ref] Membrane redox state and apoptosis: death by peroxide, Loscalzo [/bib_ref]. GPX4 is regarded as the robust central regulator and lethal signal of ferroptotic cell death, and the inactivation or absence of GPX4 can cause the accumulation of lipid peroxides. Inhibiting GPX4 will lead to a redox homeostasis imbalance, eventually resulting in ferroptotic cell death. Neomycin significantly decreased the expression of GPX4 in cochlear HCs. [bib_ref] Ferrostatin-1 protects auditory hair cells from cisplatin-induced ototoxicity in vitro and in..., Hu [/bib_ref] found that the expression of GPX4 was inhibited by cisplatin, and ROS accumulated in HEI-OC1 cells. These findings may suggest that the balance of redox homeostasis in ototoxic drug-damaged HCs is disrupted, which could promote ferroptotic cell death. Nuclear factor erythroid 2-related factor 2 (Nrf2) is a key transcription factor in the antioxidant response. When the organism is under oxidative stress, Nrf2 dissociates from Keap1 (Kelch-like ECH-associated protein 1) and binds to the antioxidant response elements of target genes such as GPX4, activating the transcription of antioxidant genes. Positively regulating the Nrf2-GPX4 axis can prevent ferroptosis in carbon tetrachloride-induced acute liver injury in mice . Knockdown of Nrf2 facilitated erastin-induced cell destruction in HepG2 cells, implying a pivotal role of Nrf2 against ferroptosis [bib_ref] Inhibition of oxidative stress and ALOX12 and NF-κB pathways contribute to the..., Dai [/bib_ref]. However, knockout of Nrf2 substantially inhibited cisplatin-induced HEI-OC1 cell death by decreasing transferrin receptor 1 protein levels and increasing GPX4 protein levels . These results suggest that the role of Nrf2 in ferroptosis regulation may be tissuespecific.
Excessive lipid peroxidation can induce ferroptosis in cells. How lipid peroxidation is generated during ferroptosis depends on the enzyme system that promotes phospholipid peroxidation. For example, ACSL4 catalyzes PUFAs in the membrane structure to produce corresponding coenzyme A, which then forms lipid peroxides through a series of downstream reactions and eventually leads to cell disintegration [bib_ref] ACSL4 dictates ferroptosis sensitivity by shaping cellular lipid composition, Doll [/bib_ref]. ACSL4 was a key enzyme in arachidonic acid-induced ferroptosis. [bib_ref] Regulation of ACSL4-catalyzed lipid peroxidation process resists cisplatin ototoxicity, He [/bib_ref] found that cisplatin-treated HCs showed significant enrichment in the arachidonic acid metabolic pathway by metabolomics assays. Researchers found that inhibiting the expression of ACSL4 could reduce the content of lipid peroxides and protect HCs [bib_ref] Regulation of ACSL4-catalyzed lipid peroxidation process resists cisplatin ototoxicity, He [/bib_ref]. These findings indicate that HCs treated with cisplatin experience lipid metabolism disorders, which, if prevented by intervention, can resist ferroptosis and reduce the destructive effect of cisplatin on HCs. Thus, ACSL4 may be a key factor contributing to cisplatin sensitivity in HCs and a potential therapeutic target for SNHL.
Several ferroptosis inhibitors may allow for potent and selective therapeutic measures for preventing HC damage. Liproxstatin-1 and ferrostatin-1, which are aromatic amine antioxidants, inhibit lipid peroxidation directly by free radical capture [bib_ref] On the Mechanism of Cytoprotection by Ferrostatin-1 and Liproxstatin-1 and the Role..., Zilka [/bib_ref]. A study demonstrated that liproxstatin-1 protected HCs from aminoglycoside-induced ototoxicity by inhibiting ferroptosis in HEI-OC1 cells and neonatal mouse cochlear HCs. Similar effects were also observed for ferrostatin-1 in damaged HCs . Regarding their mitigating effect on ototoxicity, ferroptosis inhibitors may be expected to be used to protect HCs from ototoxic drugs. In addition, from the perspective of HCs damage, ferroptosis has only been reported in ototoxic drug-induced hearing loss, and whether this pathway is involved in ARHL and NIHL needs to be further explored.
## The relationship between autophagy and ferroptosis
The relationship between autophagy and ferroptosis has attracted much attention. As universally acknowledged, autophagy is often involved in ferroptosis. Increasing evidence indicates that ferroptosis requires autophagy machinery for its execution. For example, erastin, a ferroptosis inducer, can activate autophagy, while knockout of some autophagy-related genes significantly inhibited ferroptosis induced by erastin . Autophagy can lead to the accumulation of iron ions and lipid peroxidation, eventually promoting ferroptosis.
Ferritinophagy refers to a selective autophagy process that manifests as the movement of ferritin to lysosomes and its degradation, releasing free iron. Through quantitative proteomics pair analysis of all the proteins in autophagosomes,discovered a specific protein, the nuclear receptor coactivator 4 (NCOA4), that is closely related to ferritin and so were the first to discover and name the process of ferritinophagy. Highly enriched in autolysosomes, NCOA4 mediates the targeted recognition of ferritin by autophagosomes by binding to ferritin. Ferritinophagy plays an important regulatory role in organism pathology. When ferritinophagy is over-induced, Fe 2+ overload in the cytoplasmic matrix promotes lipid peroxidation, resulting in cell membrane damage and ferroptosis [bib_ref] Ferritinophagy and ferroptosis in the management of metabolic diseases, Ajoolabady [/bib_ref]. However, blocking autophagy or knocking out NCOA4 can inhibit the accumulation of lipid ROS, so preventing the eventual occurrence of ferroptosis.
Therefore, studying the mechanism of NCOA4-mediated ferritinophagy and its pathophysiological role in different diseases may provide avenues for treatment.established a mouse model of heart failure by using the method of transverse aortic constriction. Their results showed that, compared with the control group, the accumulation of free iron and lipid peroxidation were inhibited in the hearts of mice with knockout of NCOA4. Further, the degree of left ventricular dilatation was reduced and cardiac function was improved, indicating that activating ferritinophagy can lead to the development of heart failure. Compound 9a was used to block the interaction between NCOA4 and ferritin heavy chain 1 to inhibit ferritinophagy and block ferroptosis, thus significantly improving the brain damage of rats with ischemic stroke . Similarly, cisplatin treatment promoted ROS-induced lipid peroxidation and caused iron accumulation by activating NCOA4-medicated ferritinophagy [bib_ref] Autophagydependent ferroptosis contributes to cisplatin-induced hearing loss, Jian [/bib_ref]. In summary, most studies have shown that the upregulation of ferritinophagy can promote the occurrence of ferroptosis, while the downregulation of ferritinophagy can inhibit ferroptosis. Currently, ferritinophagy is less studied in the field of auditory diseases, and targeting ferritinophagy may be a promising therapeutic strategy for the treatment of SNHL. In addition, selective autophagy processes such as lipophagy [bib_ref] Lipid storage and lipophagy regulates ferroptosis, Bai [/bib_ref] , clockophagy , and chaperone-mediated autophagy [bib_ref] Chaperonemediated autophagy is involved in the execution of ferroptosis, Wu [/bib_ref] also cause lipid peroxidation and subsequent ferroptotic cell death and may be potential targets of SNHL.
The relationship between autophagy and ferroptosis, however, is controversial. Subjecting cells to oxidative stress and injury can activate their self-protection mechanism. Autophagy may protect cells against ferroptotic cell death. Sorafenib could induce both apoptosis and ferroptosis in Desmoidtype fibromatosis cells. Furthermore, using the autophagy inhibitor hydroxychloroquine could enhance sorafenib-induced cytotoxicity. These results show that autophagy may have a pro-survival function through the inhibition of ferroptosis and apoptosis [bib_ref] In desmoid-type fibromatosis cells sorafenib induces ferroptosis and apoptosis, which are enhanced..., Schut [/bib_ref]. Similarly,reported that 15-lipoxygenase binds to PE to produce The roles of autophagy and ferroptosis in hair cell injury.
an oxidation product, 15-hydroperoxyeicosatetraenoic acid, which can lead to pro-ferroptotic cell damage and activate the pro-survival autophagy pathway to limit cell damage. Therefore, it seems to be a closely regulated mechanism between autophagy and ferroptosis. More research is needed to reveal the molecular mechanism between autophagy and ferroptosis and provide new potential therapeutic targets for hearing loss.
# Conclusion
Generally, autophagy plays a dual role in HC damage. Moderately increased autophagy can help maintain intracellular homeostasis by reducing oxidative stress, while autophagic cell death can occur under other conditions. Although many molecules that can regulate autophagy have been reported, research has been limited to external and animal experiments, and few clinical trials have examined the impact of autophagy on HCs.
Ferroptosis is widely investigated in both physiologic and pathogenic processes but has not been extensively studied in the auditory field. At present, it is reported that ototoxic drugs (e.g., neomycin and cisplatin) can induce ferroptosis in HCs, but it remains unclear whether ferroptosis engages in aging and noise-induced HC damage. Regulating ferroptosis may, therefore, be a promising strategy in SNHL therapy. However, additional studies on ferroptosis and its involvement in SNHL are needed to identify the corresponding therapeutic targets.
Increasingly, the relationship between autophagy and ferroptosis has appealed to researchers. An increasing number of studies have revealed that ferritinophagy-mediated ferroptosis is involved in the occurrence and development of neurodegenerative disease, reperfusion injury, and cancers [bib_ref] The Role of NCOA4-Mediated Ferritinophagy in Health and Disease, Santana-Codina [/bib_ref]. While ferritinophagy is a worthy target for the treatment of SNHL, its complete molecular mechanism and pathophysiological process in SNHL still call for further study. Clarification of the crosstalk between autophagy and ferroptosis would not only favor a comprehensive understanding of cell death pathways but also provide us with new ideas for the future treatment of SNHL.
# Author contributions
XC and ZH conceived and designed the study and reviewed and edited the manuscript. YS and SZ wrote the manuscript. All authors have read and approved the submitted version of the manuscript.
# Funding
This study was supported by grants from the National Natural Science Foundation of China (No. 81800915), the Fundamental Research Funds for the Central Universities (2042022kf0059), and the Key Research and Development Program of Hubei Province (2022BCA046).
## Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
## Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Frontiers in Neuroscience 06 frontiersin.org |
The role of water intake in the severity of pain and menstrual distress among females suffering from primary dysmenorrhea: a semi-experimental study
Background: Dysmenorrhea is the most common health problem among women of reproductive age. The aim of the present study was to investigate the modifying role of water intake in menstrual distress and severity of pain among young female adolescents.Method:A semi-experimental study was conducted on a sample of undergraduate female students aged 18-30 years in Isfahan, Iran from 2016 to 2019. Volunteers who had history of suffering from primary dysmenorrhea and drank less than 1600 ml water per day were assigned into water intake (n = 70) and control (n = 70) groups. Participants could select the group in which they desired to be considered. The water intake group was asked to drink water regularly based on a protocol for two menstrual periods while the control group did not receive any form of intervention. Demographic information and menstrual characteristics and severity of menstrual pain (based on a visual analogue scale), were obtained using a short questionnaire. The data were compared between and within two groups before and after intervention using chi-square test, Mann-Whitney U test, and the Friedman's analysis of variance.Results:The mean age (SD) of participants was 22.0 (2.7) years and 77 students reported normal duration of menstrual bleeding. The number of students who had normal duration of menstrual bleeding (4-6 days) in water intake group increased after intervention (39 vs. 49 after first and 46 after second cycles of menstruation). However, the interval of menstrual cycle did not change significantly in either groups. Considerable decrease in using pain killer was observed in water intake group (p < 0.001). No significant differences were observed between control and water intake groups before intervention in pain intensity (pain mean score 7.64 vs. 7.06), but within group comparison showed that pain intensity was significantly decreased among water intake group (p < 0.0001) while for control group only a significant decrease was observed for the first day of menstrual bleeding.Conclusion:The findings suggest that water intake might have modifying role in reducing menstrual bleeding duration, pain killer utilization, and pain intensity during menstrual period.
# Background
Dysmenorrhea is the most common health problem in the reproductive years of women, and it is the first reason they seek gynecological care. Dysmenorrhea is not a life-threatening disorder, but it affects women's performance in people who desire to have high and regular activities. The incidence of dysmenorrhea is reported between 16 and 93%. A study conducted in Iran showed that about 70% of young women suffer from dysmenorrhea. The symptoms of dysmenorrhea such as muscle cramp, headache, fatigue, nausea and vomiting, diarrhea and shivering can reduce the quality of life and social activities of women.
Dysmenorrhea is classified into two groups of primary and secondary dysmenorrhea. Primary dysmenorrhea is defined as a painful menstruation occurring just before or during menstruation in people who have normal ovulation and normal pelvic organs. Primary dysmenorrhea presenting with cyclic pain starts within 48 h of the first day of the menstrual cycle, and resolves by menstrual cycle day 2 or 3. Secondary dysmenorrhea is more common in 40 to 50 aged women and occurs in women with pathologic diagnosis.
Uterine activity in women with dysmenorrhea has an abnormal pattern. During menstruation, uterine endometrium produces and releases a large amount of prostaglandin F2a, etiological factor in primary dysmenorrhea, leading to uterine contraction, pain and cramps, hypoxia and ischemia of the uterus. Compared with prostaglandin F2a, Other most important vasoconstrictor agents in the nonpregnant uterus seem to be endothelin-1 and noradrenaline and arginine vasopressin (AVP). Studies have shown the important role of AVP in the onset of uterine myometrium hyperactivity and decrease in uterine blood flow, leading to appearing symptoms of early dysmenorrhea. AVP, a potent myometrial contractile agent in vitro and in vivo, seems to exert its effect on the human uterus via AVP V1a and oxytocin receptors. The effect of AVP on the smooth muscle of arteries in non-pregnant uterus may also be pronounced.
A slight deficit of water very quickly activates AVP. In low plasma concentrations, AVP induces nearly maximal renal water conservation well before activation of thirst. Data indicates that there is no difference between the plasma osmolality of people who chronically drink either low or high amounts of water, but greater vasopressin is seen in low-drinkers. Some animal studies have shown that creating dehydration in body by injection of hypertonic solution leads to the release of vasopressin. A study conducted on the impact of infusion of hypertonic saline solution in women with dysmenorrhea, showed a slight increase in hormone, leading to an increase in uterine contractions and pain. The results of a research study showed that drinking water during menstruation was associated with pain reduction and discomfort.
The prevalence of dehydration among adult people ranges from 16% to 28%and water has been considered the most popular drinking among people over the age of 6 years. The average consumption of liquids including water in Iran is approximately 1.7 and 1.9 L per day in men and women, respectively. The estimated total water intake in adults aged 20-54 years is 1307 mL/ day. This amount is 1198 mL/day in senior adults in France, and 3563 mL/day in the USA. Human water requirements should not be based on a 'minimal' intake, as this might eventually lead to a deficit and possible adverse performance and health consequences. Based on a large number of studies, daily consumption of at least 8 glasses of water can lead to the elimination of toxins and promote health.
However, the modifying role of water intake on dysmenorrhea and menstrual characteristics is not well documented. The present study aimed to investigate the role of water intake in menstrual distress and the severity of pain among females suffering primary dysmenorrhea.
# Methods
## Study design
A non-randomized semi-experimental controlled trial was conducted in Isfahan, Iran from March 2016 to March 2019. Iran National Committee for Ethics in Biomedical Research approved the study (IR.MUI. REC.1395.4). Since water drinking rate varies according to temperament, age, gender, body mass index, season, occupation and country, this study was conducted among single female university students aged 18-30 years with normal body mass index (18.5-25 kg/ m 2 ) . In order to eliminate the effect of temperature on the amount of water consumed, sampling was done during spring and autumn.
Trial registration: IRCT20180708040377N1, 16 April 2020, Retrospectively registered, at https ://www.irct.ir/trial /32446
Keywords: Primary dysmenorrhea, Pelvic pain, Arginine vasopressin, Water intake, Dehydration
## Participants
The inclusion criteria were self-reporting of pelvic pain during the first 3 days of menstruation in 3 consecutive menstrual cycle over the last 6 months, daily intake of water less than1600 ml/day and having no symptoms of vaginal discharge during the study. Individuals were asked to record the number of glasses of water they drank daily for a period of 7-10 days. Those who proved to drink less than 1600 ml/day were selected for the study. The students were excluded if they were professional athletes or if they had chronic disorders such as diabetes, cardiovascular disorders, hepatic or renal disorders, infectious disease or epilepsy, and past diagnosis of having secondary dysmenorrhea or pelvic pathology as assessed by a health professional. Students who did not desire to continue the intervention, or who followed less than 80 percent of the protocol of drinking water (1600 ml/day) also were excluded. Written informed consent was obtained from the respondents. Convenience sampling was done and participants were assigned to intervention (water intake) and control groups based on their intention. Sampling continued until the intended number of participants for each group was achieved.
## Sample size
The sample size was estimated based on pain reduction after intervention. The following formula was used to calculate the sample sizewhere the following parameters were considered: Z α/2 = 1.96 and Z β = 0.842 (for 5% significance level and power 80%); P1 = 10% (pain reduction in control group), P2 = 30% (pain reduction in water intake group); r = 1 (equal cases per each group), P = P1 + rP2/r + 1 = 20%. As such we estimated that the study would require at least 63 female student per each group. Considering about 10% drop out a sample of 70 students per each group was thought.
## Intervention
Participants in the intervention group were asked to follow a proposed protocol of drinking water per day over the two menstruation cycles. Based on this protocol, they were asked to drink a glass of water containing 250 ml 30 min before breakfast, two glasses between breakfast and lunch intermittently, one glass 30 min before lunch, two glasses between lunch and dinner intermittently, one glass 30 min after dinner, and one glass 30 min before going to bed, giving a total of 2000 ml of water intake per day. The participants in intervention group were asked to have a bottle containing 500 ml of water by themselves to be able to count the amount of water drank per day. The Control group were advised to continue drinking water based on their regular habit, and report the menstruation characteristics by completing the study questionnaire. To provide support and to prevent dropout, both groups were followed-up through social networks.
## Outcomes
The primary outcomes were recording of duration of menstrual bleeding, interval of mensuration cycle and the number of pain killer taken during menstruation. The secondary outcome was to assess pain intensity during the first 3 days of menstruation during three consequent menstrual cycles.
## Measures
In addition to demographic information both water intake and control groups were asked to record the following information in a self-designed questionnaire.
1. Duration of menstrual bleeding in days. The duration then was categorized into normal (4-6 days) and abnormal (≥ 7 days) menstruation.
[formula] n = Z α/2 + Z β 2 * P (1 − P) * (r + 1)/(P1 − P2) 2 2. [/formula]
Interval of menstruation cycles in days. The interval of menstruation was classified as short: ≤ 21 days, normal: 22-35 days, and long: ≥ 36 days. 3. The number of pain killer taken during menstruation.
This was recorded as frequency of any pharmacological pain killer taken in order to relieve pelvic pain and categorized as 0, 1-2, and 3-4. 4. The intensity of menstrual pain: This was measured using a Visual Analog Scale (VAS). The VAS is commonly used to measure menstrual pain intensity and consists of a 10-cm horizontal line with 'no pain' on one end and 'worse possible pain' on the other end. Self-reported pain is the single most important element in understanding pain.
All participants completed the questionnaire before intervention, and at the first 3 days of menstruation during three consequent menstrual cycles (before intervention, the first and the second menstrual cycle after intervention) as applied.
# Statistical analysis
The data extracted from the questionnaires were statistically analyzed using the SPSS 22. Descriptive statistics including frequency, mean and standard deviation were used to describe the data. In addition to Chi-square, t-test was used for comparison of quantitative demographic characteristics. Also duration and interval of menstruations; and the number of pain reliever taken were compared between two groups using Chi-square test. Since for some values normality was not confirmed (Kolmogorov-Smirnov test), Mann-Whitney U-test was applied to compare intensity of menstrual pain between two groups. In addition, within group comparison was performed using related samples Friedman's two-ways analysis of variance. In all instances the level of statistical significance was set at p < 0.05%.
# Results
## Characteristics of participants
In all, 140 female students were entered into the study (water intake = 70 and control group = 70). No significant differences were observed between two groups for the matched characteristics such as age, age at menarche, initiation of dysmenorrhea and menstrual regulation. The characteristics of the study samples are presented in.
## Duration of menstrual bleeding
No significant difference was observed between control and water intake groups in duration of menstrual bleeding before and after the first cycle of intervention (p = 0.094). However, the menstrual bleeding duration in intervention group significantly decreased as compared to the control group in the second cycles of menstruation (p = 0.043). The results are presented in .
## Menstruation interval
There were no significant differences between study groups before and after intervention in terms of menstruation intervals. However, number of participant in water intake group who reported normal interval of menstruation cycle increased after intervention. The control group did not experience any significant changes in menstrual cycle length at three points assessments. The results are shown in .
## Pain killer
There were no significant differences between control and water intake groups in number of pain relievers used before intervention (p = 0.066), but a significant difference was observed during the first and the second cycles of menstruation after intervention (p < 0.001). The results are shown in .
## Table 2 duration of menstrual bleeding before and after intervention in water intake and control groups
## Pain intensity
No significant differences were observed between control and water intake groups before intervention in terms of the intensity of menstruation pain during the first 3 days of menstruation (p > 0.05). However, the water intake group reported significantly less menstrual pain during the first 3 days of menstruation both in the first and the second menstrual cycle after intervention. Finally, within group comparison showed that pain intensity was significantly decreased among water intake group while for control group only a significant decrease was observed for the first day of menstrual bleeding. The results are presented in.
# Discussion
The results of this semi-experimental trial suggest that drinking 1600-2000 ml of water daily and regularly can alleviate the severity of primary dysmenorrhea, shorten the length of menstrual bleeding and reduces the average number of pharmacological pain relievers took during menstruation. Our findings concur with a previous study that revealed improvements in severity of menstrual pain in women who had water load at 90th minute after starting of dysmenorrhea. According to pain theory, arrangement of pain is cyclic and includes the following order: pain/tension/fear/ pain. Since drinking water plays a significant role Menstrual bleeding interval before and after intervention in water intake and control groups Frequency of taking pain reliever before and after intervention in water intake and control groups in reduction of vasopressin concentration, it seems that taking more water can be effective in reducing uterine contraction, diminishing tension and fear, and can be considered as a natural painkiller in the body. We found that recommended amount of water intake decreased the pelvic pain of dysmenorrhea after intervention. Furthermore, a significant improvement in menstrual pain was observed in the second cycle of intervention compared with the first cycle of intervention. This improvement could be the result of better adjustment to water intake protocol over the second cycle of intervention. Data indicated that in the first cycle of intervention, most of the participants, obeyed 80% of the protocol (consumed about 1600 ml of water per day) while the majority of the participants reached their daily water intake to 1800-2000 ml/d over the second cycle of intervention. Results also showed that intervened group consumed less pain relievers during dysmenorrhea after intervention, but their length of menstrual bleeding did not change. Participants did not encounter any specific side-effects during intervention. Two of the participants claimed positive effects of drinking water including less acne and brighter appearance. The effect of modifying water intake on the menstruation bleeding did not investigated since it needed to be considered as a single study.
The participants did not receive any follow-up assessments after intervention. Further studies with doubleblinded trial should be considered to investigate the effect of continual water intake modification on the variables in larger sample size. Future studies should also compare modifying water intake with other interventions, such as aerobic exercise and lifestyle modifications (eating behavior, regular physical activity, self-care, high-level of social relationships, and reduction of stress levels) to compare their benefits. There is evidence that risk factors such as drinking cola drinks, duration of the menstrual flow, eating meat and having a first-degree relative could affect dysmenorrhea.
This study is the first semi-experimental trial that investigated the modification role of water intake on severity of pelvic pain and menstrual distress in students with primary dysmenorrhea. We believe the findings could contribute to existing knowledge on the topic and perhaps help young women to overcome the pelvic pain of dysmenorrhea by modifying the pattern of drinking daily water. However, the data for this study was selfreported and obtained from females studying at Isfahan (Khorasgan) Azad University in Iran, which therefore restricts the extrapolation of the results to the wider populations. Thus the authors recommend other studies to be conducted in more diverse populations. In addition, one should note that we collected limited information for the study while there might be other confounding variables such as diet and alcohol consumption that potentially could influence primary dysmenorrhea. |
In Vivo Genome and Methylome Adaptation of cag-Negative Helicobacter pylori during Experimental Human Infection
Multiple studies have demonstrated rapid bacterial genome evolution during chronic infection with Helicobacter pylori. In contrast, little was known about genetic changes during the first stages of infection, when selective pressure is likely to be highest. Using single-molecule, real-time (SMRT) and Illumina sequencing technologies, we analyzed genome and methylome evolution during the first 10 weeks of infection by comparing the cag pathogenicity island (cagPAI)-negative H. pylori challenge strain BCS 100 with pairs of H. pylori reisolates from gastric antrum and corpus biopsy specimens of 10 human volunteers who had been infected with this strain as part of a vaccine trial. Most genetic changes detected in the reisolates affected genes with a surface-related role or a predicted function in peptide uptake. Apart from phenotypic changes of the bacterial envelope, a duplication of the catalase gene was observed in one reisolate, which resulted in higher catalase activity and improved survival under oxidative stress conditions. The methylomes also varied in some of the reisolates, mostly by activity switching of phase-variable methyltransferase (MTase) genes. The observed in vivo mutation spectrum was remarkable for a very high proportion of nonsynonymous mutations. Although the data showed substantial within-strain genome diversity in the challenge strain, most antrum and corpus reisolates from the same volunteers were highly similar to each other, indicating that the challenge infection represents a major selective bottleneck shaping the transmitted population. Our findings suggest rapid in vivo selection of H. pylori during early-phase infection providing adaptation to different individuals by common mechanisms of genetic and epigenetic alterations. IMPORTANCE Exceptional genetic diversity and variability are hallmarks of Helicobacter pylori, but the biological role of this plasticity remains incompletely understood. Here, we had the rare opportunity to investigate the molecular evolution during the first weeks of H. pylori infection by comparing the genomes and epigenomes of H. pylori strain BCS 100 used to challenge human volunteers in a vaccine trial with those of bacteria reisolated from the volunteers 10 weeks after the challenge. The data provide molecular insights into the process of establishment of this highly versatile pathogen in 10 different human individual hosts, showing, for example, selection for changes in hostinteraction molecules as well as changes in epigenetic methylation patterns. The data provide important clues to the early adaptation of H. pylori to new host niches after transmission, which we believe is vital to understand its success as a chronic pathogen and develop more efficient treatments and vaccines.
H elicobacter pylori is a highly prevalent human pathogen that causes chronic inflammation of the stomach epithelium. Although the majority of infected individuals do not develop further symptoms, H. pylori infection can give rise to clinical disease, including gastroduodenal ulcers, lymphoma of the mucosa-associated lymphoid tissue (MALT), and gastric cancer [bib_ref] Helicobacter pylori infection, Suerbaum [/bib_ref]. H. pylori has been recognized as a class I carcinogenic agent by the World Health Organization (WHO) since 1994. If not treated, this human gastric pathogen usually establishes a lifelong infection [bib_ref] Helicobacter pylori persistence: biology and disease, Blaser [/bib_ref]. Transmission mostly takes place during childhood due to direct contact with infected relatives, but the infection is typically diagnosed during adult life [bib_ref] Epidemiology of Helicobacter pylori infection, Burucoa [/bib_ref].
A successful colonization of the stomach niche requires, among many other traits, urease activity [bib_ref] Gastric infection by Helicobacter pylori, Sachs [/bib_ref] , the ability of H. pylori to swim in the mucus layer using flagellumbased chemotactic motility [bib_ref] Chemotaxis plays multiple roles during Helicobacter pylori animal infection, Terry [/bib_ref] , and multiple adhesins to attach to the gastric epithelium [bib_ref] Helicobacter pylori infection: an overview of bacterial virulence factors and pathogenesis, Kao [/bib_ref] [bib_ref] Factors that mediate colonization of the human stomach by Helicobacter pylori, Dunne [/bib_ref].
Genome studies of sequential H. pylori isolates from chronically infected individuals demonstrated that H. pylori undergoes rapid in vivo genome evolution [bib_ref] Recombination and mutation during long-term gastric colonization by Helicobacter pylori: estimates of..., Falush [/bib_ref] [bib_ref] Microevolution of Helicobacter pylori during prolonged infection of single hosts and within..., Morelli [/bib_ref] [bib_ref] Bidirectional genomic exchange between Helicobacter pylori strains from a family in Coventry,..., Krebes [/bib_ref]. This is in part a result of the natural competence of this bacterium, which allows it to take up extracellular DNA, leading to chromosome exchange during mixed infection with other H. pylori strains meeting in the same stomach niche. Moreover, a high mutation rate contributes to an elevated genetic diversity [bib_ref] Mutation frequency and biological cost of antibiotic resistance in Helicobacter pylori, Bjorkholm [/bib_ref]. The high mutation rate results from the lack of genes involved in a traditional mismatch repair (MMR) system [bib_ref] Bacterial populations as perfect gases: genomic integrity and diversification tensions in Helicobacter..., Kang [/bib_ref] [bib_ref] Recombination and DNA repair in Helicobacter pylori, Dorer [/bib_ref] and mutagenic properties of the H. pylori polymerase I (Pol I) enzyme [bib_ref] Unexpected role for Helicobacter pylori DNA polymerase I as a source of..., Garcia-Ortiz [/bib_ref]. Genes with an outer membrane-related role vary at an elevated apparent rate, probably due to positive selection [bib_ref] Bidirectional genomic exchange between Helicobacter pylori strains from a family in Coventry,..., Krebes [/bib_ref] [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref] , reinforcing the idea that genes involved in adhesion and host interaction play a vital role in establishing and maintaining the chronic infection.
Furthermore, H. pylori has a relatively small genome but encodes a large number of restriction-modification (R-M) systems. Every H. pylori strain carries a unique set of R-M systems, leading to variable methylomes [bib_ref] Diverse functions of restriction-modification systems in addition to cellular defense, Vasu [/bib_ref]. It was recently verified that R-M systems act as barriers against heterologous incoming DNA (e.g., antibiotic resistance cassettes), while restriction endonucleases (REases) do not limit the import of homeologous DNA from other H. pylori strains [bib_ref] Genome-wide analysis of chromosomal import patterns after natural transformation of Helicobacter pylori, Bubendorfer [/bib_ref]. In addition to the role of methyltransferases (MTases) in self-DNA protection as parts of R-M systems, methylation controls several bacterial functions by regulating the expression of genes. Many MTase genes include repetitive nucleotide tracts prone to phase variation by slipped strand mispairing. It has been suggested that phase-variable MTases might play a role in the adaptation of H. pylori and other pathogens to new hosts due to the regulation of the expression of genes involved in colonization and pathogenesis [bib_ref] Haemophilus influenzae phasevarions have evolved from type III DNA restriction systems into..., Fox [/bib_ref] [bib_ref] Phasevarions mediate random switching of gene expression in pathogenic Neisseria, Srikhanta [/bib_ref] [bib_ref] Phasevarion mediated epigenetic gene regulation in Helicobacter pylori, Srikhanta [/bib_ref].
So far, little is known about the genome and especially the methylome evolution of H. pylori during the early phases of the infection and the role of these changes in the pathogen's adaptation to different stomach conditions. How H. pylori genomic diversity influences the ability of the bacteria to colonize and survive in a new stomach niche is not well understood. To our knowledge, only one previous study from our group has explored whole-genome and methylome adaptation to new stomach niches during an experimental human infection. In that prior study [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref] , the H. pylori strain BCM-300 was used to challenge human volunteers as part of a vaccine trial. BCM-300 carries a functional cag pathogenicity island (cagPAI), generating higher levels of inflammation. However, only one H. pylori reisolate per individual was available from that study, so that we could not study heterogeneity of the H. pylori population within one stomach during this early-stage infection.
In the present study, we used single-molecule, real-time (SMRT) sequencing technology to obtain and compare finished closed genome sequences and methylomes of the challenge strain and pairs of H. pylori reisolates from the antrum and corpus of 10 volunteers who participated in an experimental infection study performed with the cagPAI-negative challenge strain BCS 100 [bib_ref] Correlation of T cell response and bacterial clearance in human volunteers challenged..., Aebischer [/bib_ref] [bib_ref] Challenge model for Helicobacter pylori infection in human volunteers, Graham [/bib_ref] , with the aim of detecting genomic and epigenomic alterations during early infection. In addition, we compared the genome sequences of 16 purified single colonies from the parental (challenge) strain to assess the homogeneity of the inoculum. Whole-genome and methylome comparisons revealed substantial genomic diversification during early human infection affecting outer membrane-related genes and, interestingly, genes involved in peptide uptake. In addition to nucleotide sequence changes, switching of the activity of phase-variable MTases generated methylome variability between reisolates, pointing to a role of methylation in adaptation via altering the expression of genes.
# Results
Genome analysis of the BCS 100 challenge strain. The cagPAI-negative H. pylori strain BCS 100 was originally isolated from a patient with mild superficial gastritis and was the first H. pylori strain used for a challenge study in human volunteers [bib_ref] Challenge model for Helicobacter pylori infection in human volunteers, Graham [/bib_ref]. In contrast to most H. pylori strains used for research, BCS 100 had not been subjected to purification from single colonies, in order to preserve its full infectivity. Subclones H1 to H16 were subsequently purified from BCS 100 single colonies, and a draft genome sequence of the H1 clone had already been obtained previously, by 454 sequencing [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref]. In order to be able to understand the dynamics of genome and methylome adaptation during early-stage human infection with BCS 100, it was important to initially assess the within-population diversity of the challenge strain. We used SMRT sequencing technology to obtain the complete genome sequence and the methylome of one single colony-purified clone (H1) of BCS 100. In addition, we used Illumina MiSeq technology to obtain draft genomes from the 15 other clones (H2 to H16). In total, we analyzed 16 genomes from the BCS 100 input strain population. Among clones H1 to H16, single nucleotide polymorphisms (SNPs) were identified at 16 positions (see [fig_ref] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge... [/fig_ref] in the supplemental material). Individual clones H2 to H16 differed from the reference clone H1 at one to five positions each, and none of the clones H2 to H16 was identical to H1 at all 16 polymorphic positions. Eighty percent of the polymorphic positions were nonsynonymous. In addition to the single nucleotide polymorphisms, the comparisons between H1 and H2 to H16 identified four clusters of nucleotide polymorphisms (CNPs; see Materials and Methods for definition) in four different clones [fig_ref] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge... [/fig_ref]. The four CNPs and five out of the 16 SNPs were located within a specificity subunit of a predicted type I R-M system with yet-unknown target motif and activity. The MTase of this R-M system displays more than 90% nucleotide identity with M.HpyAXIII from the H. pylori strain 26695 [bib_ref] The complete genome sequence of the gastric pathogen Helicobacter pylori, Tomb [/bib_ref]. The type I R-M system consists of one MTase gene, one REase gene, and two S subunits. The two S subunits are almost identical in sequence and located in two different genomic locations. This situation could be the result either of a recent duplication or of concerted evolution [bib_ref] Concerted evolution between duplicated genetic elements in Helicobacter pylori, Pride [/bib_ref]. All the polymorphisms found in the study affected only the second S subunit in both loci which in one locus was located between the first S subunit and the REase of the type I R-M system (position in H1: 841163 to 841921) and in the other location between the outer membrane protein (OMP) gene hofF and the duplicated S subunit (position in H1: 786359 to 787114). The four CNPs did not share any polymorphisms. This overall small number of polymorphisms indicated a low level of heterogeneity within the challenge strain population, consistent with known patterns of within-host strain diversity [bib_ref] Within-host evolution of Helicobacter pylori shaped by niche-specific adaptation, intragastric migrations and..., Ailloud [/bib_ref]. No evidence of an ongoing mixed infection with unrelated H. pylori strains was obtained.
Whole-genome comparison of the challenge strain and reisolates. In order to assess genome and methylome changes during early-stage infection of human volunteers with H. pylori BCS 100, pairs of H. pylori reisolates were recovered from antrum and corpus biopsy specimens obtained from the 10 human volunteers from the challenge group of the vaccine study (five each randomly selected from the vaccine and control arms of the study), at 10 weeks after H. pylori challenge [bib_ref] Correlation of T cell response and bacterial clearance in human volunteers challenged..., Aebischer [/bib_ref]. For all 20 reisolates, finished closed genome sequences including methylome data were obtained using SMRT sequencing technology [fig_ref] FIG 1: A [/fig_ref].
We first studied how the intrinsic diversity observed in clones H1 to H16 of BCS 100 was reflected in the reisolates. Eleven of the 20 reisolates were identical to H1 at all 16 SNP positions. Most (13/16) of the single polymorphic nucleotides identified among clones H1 to H16 were not polymorphic among the reisolates, which all were identical to H1 at these positions. Only three of the polymorphisms detected among the inoculum clones were also detected in nine reisolates, in different combinations. Notably, a combination of two nonsynonymous SNPs in clone H3, affecting a predicted multidrug efflux transporter and the pdxJ gene, was also found in six reisolates (8A3, 8C10, 29A2, 29C8, 125A2, and 125C7), suggesting that those reisolates and clone H3 derive from a recent common ancestor. A nonsynonymous SNP in the copA gene, found in H2 and H9 of the input population, was also present in reisolate 119A2 [fig_ref] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge... [/fig_ref].
Since sequencing 16 clones from the inoculum may not fully capture the inherent population variability, we next analyzed genetic variants that were present in the reisolates but not detected in any of clones H1 to H16. We observed 86 SNPs shared among reisolates from different volunteers, here called non-unique SNPs. Eighty-three non-unique SNPs were present in the same group of reisolates (12A3, 12C8, 81A1, 81C9, and 87C7), and three were present in the same group of reisolates except 87C7. In addition, we detected six CNPs shared between these five reisolates and one CNP shared between 12A3, 12C8, 81A1, and 81C9, many of which were within the outer [bib_ref] Correlation of T cell response and bacterial clearance in human volunteers challenged..., Aebischer [/bib_ref] which was the basis of the present study. Human volunteers were given a Salmonella Ty21a live vaccine control or a recombinant Ty21a vaccine expressing the H. pylori urease. Forty-two days later, human volunteers were challenged with the cagPAI-negative BCS 100 H. pylori strain, a heterogeneous strain from which several subclones (H1 to H16) were genome sequenced. Six and 10 weeks postchallenge (wpc), gastroscopies were performed and blood samples were collected. Reisolates were cultured from antrum and corpus biopsy specimens taken 10 wpc, and reisolates from 10 volunteers were subjected to SMRT sequencing technology to obtain complete genome sequences and methylome data. (B) Phylogenetic tree representing the genomes of all strains. The reisolates 12A3, 12C8, 81A1, 81C9, and 87C7 represent a distinct subpopulation compared to the clones obtained from the challenge strain H1 to H16. Blue, H1 to H16 clones. Red, volunteers infected by reisolates descending from distinct subpopulations of the inoculum. The scale bar indicates substitutions per site. Statistical branch support is indicated by posterior probability values and node circles.
## Estibariz et al.
® membrane protein (OMP)-encoding genes babA, babB, hopQ, and hofC. One CNP was located within the S subunit of a putative type I R-M system with unknown target sequence and activity status, and one CNP was located in an intergenic region [fig_ref] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge... [/fig_ref]. The data are most compatible with an in vivo selection of a subpopulation that was likely present in the heterogeneous input strain population BCS 100 but not represented by clones H1 to H16. Reisolates 12A3, 12C8, 81A1, 81C9, and 87C7 might have all evolved from this precursor. In agreement with this, these reisolates all cluster together in a phylogenetic tree [fig_ref] FIG 1: A [/fig_ref]. We note that most antrum and corpus reisolates from the same volunteer were very closely related and more similar to each other than to any other strain in the tree [fig_ref] FIG 1: A [/fig_ref]. Only a few reisolate clones, including paired antrum and corpus clones from two volunteers, clustered apart from each other and from the inoculum diversity sampled in H1 to H16 [fig_ref] FIG 1: A [/fig_ref]. The common clustering of reisolates from most volunteers suggests a colonization by very similar lineages selected from the heterogeneous inoculum.
We next investigated unique polymorphisms present in either one single reisolate (unique SNPs) or the two paired reisolates collected from the same volunteer (pairspecific SNPs). Those genetic modifications are most likely to have emerged de novo, during the short duration of the challenge infection (10 weeks). Whole-genome analysis of all 20 reisolates versus the 16 colonies from the challenge strain revealed a total of 25 SNPs that comprised 23 unique SNPs, and 2 pair-specific SNPs [fig_ref] TABLE 1: Key characteristics of the genome sequences of H [/fig_ref]. We also detected one unique CNP in one reisolate [fig_ref] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge... [/fig_ref]. The number of unique SNPs per reisolate varied between 0 and 3. The average mutation rate was 4.5 ϫ 10 Ϫ6 mutations per site per year. The mutation rates for the antrum and corpus reisolates were 5.0 ϫ 10 Ϫ6 and 4.0 ϫ 10 Ϫ6 mutations per site per year, respectively. The two pairspecific SNPs were excluded from this analysis, since they were present in both antrum and corpus reisolates from one infected volunteer, and they are therefore likely to already have been present in the inoculum, despite not being represented in clones H1 to H16. The mutation rates for control and vaccination groups were 5.0 ϫ 10 Ϫ6 and 4.0 ϫ 10 Ϫ6 mutations per site per year [fig_ref] TABLE 1: Key characteristics of the genome sequences of H [/fig_ref] , respectively. Statistical analysis of the mutation frequencies (unpaired t test) did not show a significant difference between the mutation rates (antrum versus corpus, P ϭ 0.4566, and vaccinated versus control, P ϭ 0.4585). Taken together, the mutation rates calculated here were in agreement with previous studies [bib_ref] Microevolution of Helicobacter pylori during prolonged infection of single hosts and within..., Morelli [/bib_ref] [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref]. A very large proportion of unique SNPs (76%) were nonsynonymous, probably resulting from a combination of diversifying selection and a lack of time to purge slightly deleterious nonsynonymous mutations (see Discussion).
Most unique SNPs were detected in genes belonging to the functional category of "Cell envelope, transport and binding proteins" (genolist.pasteur.fr/PyloriGene/) [fig_ref] TABLE 2: Unique and pair-specific SNPs in H [/fig_ref]. In the subsequent paragraphs, we will highlight a selection of observed changes. Intragenomic rearrangements affecting catalase and lipopolysaccharide (LPS) biosynthesis genes. The genome of the reisolate 103C8 contained an insertion of 3,433 bp that was initially identified by SMRT sequencing and then confirmed by PCR. The inserted 3-kb fragment contained a second copy of the catalase gene (katA, hp0875, Rapid Annotation Server [RAST]: fig|210.1715.peg.852) and a fusion of duplicated fragments of the genes frpB and kapA [fig_ref] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and... [/fig_ref]. Since catalase plays an important role in the detoxification of oxygen radicals, we tested whether this duplication had phenotypical effects. Catalase gene transcript was measured by quantitative PCR (qPCR), and it was significantly higher in the reisolate 103C8 than the wild-type strain [fig_ref] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and... [/fig_ref]. We measured catalase activity in the respective bacterial lysates, demonstrating that catalase activity in the reisolate 103C8 with two katA copies was approximately twice that of the challenge strain [fig_ref] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and... [/fig_ref]. We hypothesized that increased catalase activity might help with the detoxification of oxygen radicals and serve to defend against oxidative stress. Thus, the sensitivity to oxidative stress was also comparatively tested for both strains, using paraquat (PQ) as oxidizing agent. The subclone H1 displayed a significant growth delay when exposed to 10 M PQ, while growth of the 103C8 reisolate was unaffected under the same conditions [fig_ref] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and... [/fig_ref]. This result strongly suggests that a second katA copy confers a selective advantage to the reisolate 103C8 under conditions of oxidative stress. Genome analysis of the reisolate 29C8 identified a CNP within a gene encoding an LPS biosynthesis glycosyltransferase (jhp0562 homolog). This glycosyltransferase was previously shown to be involved in type 1 and type 2 Lewis antigen synthesis [bib_ref] Novel functions for the glycosyltransferases Jhp0562 and GalT in Lewis antigen synthesis..., Pohl [/bib_ref]. Moreover, the presence of this gene was positively associated with the presence of several virulence factors including the cagPAI, vacA, and several adhesins and has been linked with a higher incidence of peptic ulcers in children [bib_ref] Clinical relevance and diversity of two homologous genes encoding glycosyltransferases in Helicobacter..., Oleastro [/bib_ref]. The CNP included six synonymous and two nonsynonymous SNPs compared with H1 to H16. Since a recombination with an unrelated strain of H. pylori was quite unlikely in the short-term infection experiment, we performed a genome alignment between the jhp0562 homolog and its paralogous downstream gene, coding for the -(1,3)-galactosyltransferase [-(1,3) galT]. The CNP sequence in the jhp0562 homolog was identical to the corresponding region in the paralog galT (jhp0563), suggesting that this CNP was generated by an intragenomic recombination event [fig_ref] FIG 1: A [/fig_ref] and B). Intragenomic rearrangements between these two genes were in fact described previously [bib_ref] Novel functions for the glycosyltransferases Jhp0562 and GalT in Lewis antigen synthesis..., Pohl [/bib_ref].
Diversification of the Opp oligopeptide ABC transporter system. The oligopeptide transport system (Opp) is typically encoded by five genes in Gram-negative bacteria: the periplasmic oligopeptide-binding protein (oppA), two permeases (oppB and oppC), and two ATP-binding subunits (oppD and oppF) [bib_ref] Neisseria meningitidis is structured in clades associated with restriction modification systems that..., Budroni [/bib_ref]. In H. pylori, the genes belonging to the Opp system are located in two different genomic clusters, oppA-oppB and oppC-oppD. No homolog for the oppF gene has been identified until now [fig_ref] FIG 3: Graphical representation of the H [/fig_ref]. Transcription of the catalase gene in various H. pylori clones was measured by qPCR. katA transcript is shown as % relative to the H1 reference, which was set to 100%. The reisolate 103C8 with two katA copies showed significantly higher katA transcript amounts than H1. A mutant lacking the katA gene did not show transcript, as expected. All results were normalized to the respective 16S transcript amounts, also using strain H1 as reference. (C) Catalase activity was measured in bacterial lysates of strain H1, the reisolate 103C8, and H1ΔkatA (negative control) The reisolate 103C8, with two katA copies, displayed higher catalase activity than H1. The calculation of the catalase units/ml is defined as in the Megazyme catalase assay kit instructions. Test, one-way analysis of variance (ANOVA) (P Ͻ 0.05). (D) Survival experiment using paraquat (PQ) as oxidizing agent on live bacteria. Results are shown as relative survival values, where individual data sets were normalized to their respective bacterial counts obtained at time zero hours, which were set to 1. Strains H1 and 103C8 were treated with 10 M PQ or left untreated (untr), and the number of colonies was counted 10 h postexposure. Strain H1 was less able to resist the oxidative stress, in contrast to reisolate 103C8, which was more resistant to PQ. Test, two-way ANOVA (P Ͻ 0.05). For panel B, dotted lines refer to the duplicated and fused regions from the kapA and frpB1 genes. For panels C and D, * ϭ P Ͻ 0.05; ** ϭ P Ͻ 0.01; ns, not significant.
H. pylori Evolution during Early-Stage Human Infection ® Comparing the opp loci between reisolates and challenge strain, we discovered four nonsynonymous SNPs and one SNP introducing a stop codon affecting the genes of the ABC transporter for oligopeptides in four different reisolates. Frameshift mutations leading to a truncation of one of the open reading frames were observed in two more reisolates and [fig_ref] FIG 3: Graphical representation of the H [/fig_ref]. The gene oppD in H1 contains a homopolymeric tract of six adenine residues, leading to a truncated protein. In contrast, the H2 to H16 clones of the challenge strain and all 20 reisolates had seven adenines within the same homopolymeric tract, resulting in a predicted full-length OppD protein. The oppD gene of reisolate 78A3 contained a premature stop codon unrelated to the homopolymeric tract. Moreover, we observed several modifications affecting the opp genes within the BCS 100 strain, including insertions of 57 bp in the oppD gene of H4 and 633 bp within the oppA gene of H16, which caused the truncation of the reading frame in H16. Using a combination of primers, we found two differently sized bands when the oppD and oppA genes were amplified by PCR in clones H4 and H16 (data not shown). We picked 25 colonies from H4 and H16, and using the same combination of primers, we observed that single colonies carried either the insertion or the wild-type allele. Thus, we found that clones H4 and H16 of the BCS 100 challenge strain had developed heterogeneous oppD and oppA loci due to an unknown mechanism, despite being the result of single colony purification.
In order to evaluate the species-wide variation in genotype of the opp gene clusters, we extracted the opp gene sequences of 75 representative H. pylori whole-genome sequences from the NCBI database. The four genes were present in all 75 H. pylori strains. Based on the gene sequence, all four opp genes were predicted to be functional in 58/75 strains. Putatively inactive oppA, oppC, and oppD genes were observed in 13.33%, 9.33%, and 2.66% of strains, respectively . In contrast, the gene sequence of oppB was complete in all the genomes, suggesting an important and conserved role of this gene.
Selective pressure on the babA adhesin gene during early-stage human infection. Several studies have shown that OMP-encoding genes exhibit sequence changes at higher frequencies than housekeeping genes during the course of H. pylori infection [bib_ref] Bidirectional genomic exchange between Helicobacter pylori strains from a family in Coventry,..., Krebes [/bib_ref] [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] [bib_ref] Helicobacter pylori genomic microevolution during naturally occurring transmission between adults, Linz [/bib_ref]. In line with these previous observations, we observed several non-unique CNPs and SNPs affecting OMP-related genes. One of the genes most affected by genomic changes was babA, the gene encoding the well-characterized adhesin binding . Three reisolates carried unique nonsynonymous SNPs or a premature stop codon in babA. In addition, a deletion of approximately 1 kb was found in reisolate 29C8 [fig_ref] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and... [/fig_ref]. Although BabA expression was confirmed by Western blotting in H1, a Le b -binding assay showed that the H1 strain and a representative subset of the reisolates (8A3, 8C10, 12A3, 12C8, 29A2, 29C8, 48A2, and 48C8) were not able to bind Le b compared to the strain J99, which is capable of binding [fig_ref] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and... [/fig_ref].
Complete methylome analysis of challenge strain BCS 100 clone H1 and reisolates. Every H. pylori strain carries a large number of active R-M systems that differ from the R-M set of other strains, leading to strain-specific methylomes [bib_ref] Restrictionmodification systems may be associated with Helicobacter pylori virulence, Ando [/bib_ref] [bib_ref] Identification of type II restriction and modification systems in Helicobacter pylori reveals..., Xu [/bib_ref] [bib_ref] The complex methylome of the human gastric pathogen Helicobacter pylori, Krebes [/bib_ref]. The plasticity of the methylome during early-phase infection of a new host is still not well understood.
We comparatively analyzed the methylomes of reference challenge clone H1 and all the reisolates using SMRT sequencing technology. In total, 20 different methylated motifs were detected in our set of sequences. Nineteen of these were detected as methylated in H1, and 18 motifs were methylated in H1 and all the reisolates [fig_ref] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge... [/fig_ref] and . The methylation status of the remaining two motifs (CY m6 AN 6 TRG and CCA m6 AK) varied between isolates, suggesting phase variation of the responsible MTase genes. The CYAN 6 TRG motif was methylated in H1 and in 14 of the reisolates. This motif was not detected as methylated in reisolates 12A3, 12C8, 48C8, 81A1, 81C9, and 87C7, the group of isolates that derive from a common ancestor not represented by clones H1 to H16. The motif CCAAK was detected as methylated in only four reisolates (12A3, 12C8, 81A1, and 81C9) and not in H1 or the rest of the reisolates.
Using the REBASE database [bib_ref] REBASE-a database for DNA restriction and modification: enzymes, genes and genomes, Roberts [/bib_ref] and homology-based prediction analysis, we were able to assign 16 motifs to already-known H. pylori MTases . While m6 A and m4 C methylation can be detected by SMRT sequencing, a similarly reliable detection of m5 C modifications by SMRT sequencing is not possible [bib_ref] Enhanced 5-methylcytosine detection in single-molecule, realtime sequencing via Tet1 oxidation, Clark [/bib_ref]. The genome of H1 contains
[formula] ϩ Ϫ Ϫ ϩ Ϫ ϩ Ϫ Ϫ Ϫ ATTAAT m6 A 99.53 856 ϩ ϩ ϩ ϩ ϩ ϩ ϩ ϩ ϩ GTAC m6 A 100 198 ϩ ϩ ϩ ϩ ϩ ϩ ϩ ϩ ϩ GGCAA m6 A 100 3,364 ϩ ϩ ϩ ϩ ϩ ϩ ϩ ϩ ϩ CCAAK m6 A 99.95 6,357 Ϫ ϩ ϩ Ϫ Ϫ Ϫ ϩ ϩ Ϫ *CGCGCNY m4 C Ϫ ϩ Ϫ ϩ Ϫ ϩ Ϫ Ϫ Ϫ **TGCAGA m6 A Ϫ Ϫ Ϫ ϩ Ϫ Ϫ Ϫ Ϫ Ϫ a [/formula]
The % of motifs detected and the total number of motifs in the genome are based on the methylome of clone H1. The CCAAK motif quantitation is based on the 12C8 sequence, since the motif was not methylated in clone H1. ϩ means methylation, Ϫ means absence of methylation, novel motifs are indicated in italic, and phase-variable MTases are shaded in gray. Modified bases are in bold. Underlined bases refer to the modified base in the complementary strand. *, the GCGC motif that cannot be reliably detected with SMRT sequencing. **, the motif TGCAGA was found only in 29C8 (62.78% of the motifs detected, 309 motifs within the genome), and it might probably be the motif TGCA. #, the following reisolates had methylation patterns identical to strain H1: 8A3, 8C10, 29A2, 48A2, 78C8, 87A3, 103A4, 103C8, 119A2, 119C10, 125A3, and 125C7.
genes coding for homologs of three known m5 C MTases in H. pylori, and we used restriction analysis using HhaI, HaeIII, and HpyCH4IV restriction enzymes to test the methylation status of predicted m5 C motifs. The data confirmed methylation of the G m5 CGC, GG m5 CC, and A m5 CGT motifs in the genome of H1 and likely in the rest of the reisolates since there were no differences in the relevant MTase sequences [fig_ref] FIG 3: Graphical representation of the H [/fig_ref]. The CCTC and the G m6 AGG motifs are methylated by two MTases that belong to the same type IIS system, one m5 C and one m6 A MTase, respectively. The G m6 AGG motif was detected as methylated in H1 [fig_ref] FIG 3: Graphical representation of the H [/fig_ref] and all reisolates, and the fact that both MTase genes of this R-M system are intact (i.e., do not contain premature stop codons) suggests that the motif CCTC is methylated as well. Accordingly, we identified 24 methylated motifs corresponding to 22 active type II and type III R-M systems [fig_ref] FIG 3: Graphical representation of the H [/fig_ref] , and . As stated above, the BCS 100 genome contains a predicted type I R-M system that contained multiple sequence polymorphisms, but its target motif and activity status remain unknown. Based on the gene sequences, clones H2 to H16 of the challenge strain are predicted to have the same active MTases as reference clone H1.
Two novel motifs that had not been described before (CCA m6 AK and GCRC m6 A) and two motifs that had not yet been assigned to MTase genes but had been detected previously in other H. pylori strains, GGCA m6 A (44) and VCGR m6 AG (the motif appears as methylated in the REBASE), were found in this study. Inactivation of selected ® candidate genes (classified as "genes with unknown function" in our RAST annotation) via insertion of an antibiotic cassette and subsequent SMRT sequencing allowed us to discover three novel MTase genes responsible for the methylation of CCA m6 AK, GCRC m6 A, and GGCA m6 A, respectively. Complete loss of methylation occurred in the mutants with inactivated MTases. None of the other candidate genes tested were responsible for the methylation of VCGR m6 AG. The MTase responsible for the methylation of this motif thus remains unknown.
The novel MTase (M.HpyH1I) methylating CCA m6 AK is a phase-variable type III enzyme that contains one homopolymeric tract with 13 guanine residues in frame. The motif was methylated only in the reisolates 12A3, 12C8, 81A1, and 81C9. Sanger sequencing was performed to confirm the number of guanines for all strains. The four reisolates containing methylated CCAAK motifs had an intact open reading frame, while all other reisolates had either eight, nine, or 12 guanine residues, resulting in split open reading frames [fig_ref] FIG 4: Circos plot displaying the distribution of methylated sequence motifs in H [/fig_ref].
A second phase-variable R-M system (HpyH1III) methylating the motif CY m6 AN 6 TRG was found. The MTase in charge was identified by homology. This motif was methylated in 14 reisolates and in reference clone H1 but not in the reisolates 12A3, 12C8, 48C8, 81A1, 81C9, and 87C7. The status of the homopolymeric tract was checked by Sanger sequencing and was in complete agreement with the methylation patterns in all strains [fig_ref] FIG 4: Circos plot displaying the distribution of methylated sequence motifs in H [/fig_ref]. One homologous enzyme has been described in H. pylori UM032 (44). Lee and colleagues discovered that a 12-guanine homopolymeric tract split the S subunit in two (S1 and S2), and the expression of S1 led to methylation of the motif.
# Discussion
Due to its high mutation and recombination rates and the resulting very high level of within-species genetic diversity, H. pylori has become a paradigm for the within-host evolution of bacterial pathogens. Most studies of the in vivo evolution of the H. pylori genome were performed with sequential isolates from chronically infected adult patients. In contrast, little is yet known about the genome evolution of H. pylori during the initial phase of the infection, when selection is likely to be strongest due to the need for the bacteria to establish infection in a new host after a massive bottleneck caused by the transmission [bib_ref] Microevolution of Helicobacter pylori during prolonged infection of single hosts and within..., Morelli [/bib_ref] [bib_ref] Genomic evolution and transmission of Helicobacter pylori in two South African families, Didelot [/bib_ref] [bib_ref] Genomic changes during chronic Helicobacter pylori infection, Kraft [/bib_ref]. Furthermore, the methylomes of H. pylori are highly complex, and so far, the role of epigenetic modifications in the adaptation of this gastric pathogen to a new stomach niche or new hosts is not known. To our knowledge, only one previous study from our group has described the in vivo methylome adaptation during acute infection, where methylome variation during infection was observed as a result of phase variability in MTase genes [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref]. In the present study, we report the genome and methylome evolution of H. pylori after a short-term infection of human volunteers with a cagPAI-negative strain, BCS 100. In contrast to the previous study, reisolates from two different stomach locations from each volunteer were available for analysis.
All H. pylori strains cultured from unrelated individuals display specific genotypic and phenotypic characteristics. There is also extensive intrastrain heterogeneity, such that H. pylori has been termed a "quasispecies" [bib_ref] Quasispecies development of Helicobacter pylori observed in paired isolates obtained years apart..., Kuipers [/bib_ref]. Human volunteer infection studies are particularly informative about genome-based adaptation to a new host, since the duration of infection is precisely known, and both ancestral and evolved strains can be characterized in depth. In the present work, the analysis of 16 purified single colonies from the challenge strain BCS 100 showed that it is not genetically homogeneous. In addition to differences among the 16 clones purified from the challenge strain population, intraclonal variability within some of the single colony-purified clones of the BCS 100 strain was detected, including insertions and deletions within opp genes. Our observation of genetic heterogeneity within BCS 100 is in agreement with a previous study [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] , where diversity at six loci selected from a draft genome comparison between clone H1 and a single reisolate (8A3) corresponded to nucleotide polymorphisms now found to be present in clones H1 to H16. In addition, the observation of several non-unique SNPs and CNPs that were detected in multiple reisolates but in none of the clones H1 to H16 provides strong evidence that the challenge strain contained further subpopulations not sampled in clones H1 to H16. Despite the diversity observed in the inoculum, phylogenetic analysis of the challenge strain and the reisolates [fig_ref] FIG 1: A [/fig_ref] revealed that eight out of 10 volunteers were likely colonized by the same subpopulation from the inoculum. In two volunteers, paired isolates from the antrum and corpus belonged to distinct subclusters. Both these observations imply that H. pylori populations experience a strong bottleneck after person-to-person transmission. Previous experimental infection studies in humans [bib_ref] Challenge model for Helicobacter pylori infection in human volunteers, Graham [/bib_ref] and in mice [bib_ref] High-resolution mapping reveals that microniches in the gastric glands control Helicobacter pylori..., Fung [/bib_ref] suggested that only a small number of bacterial cells are establishing the infection early on. Our data indicate that these initial colonizing clones are also likely to be genetically homogenous, with few exceptions. Consequently, the distinct selective environmental pressures of a new gastric niche might result in a strong reduction of the inoculum diversity during the early stages of infection.
Whole-genome comparisons of the reisolates with the challenge strain allowed us to calculate in vivo mutation rates. The calculated mutation rates from our data were in agreement with estimates obtained in previous studies that were performed with sequential isolates from chronically infected individuals, and with one previous study performed on isolates from challenged human volunteers [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref] [bib_ref] Microevolution of virulence-related genes in Helicobacter pylori familial infection, Furuta [/bib_ref]. Notably, this is the second independent study performed on reisolates from challenged human volunteers that did not show evidence of a "mutational burst" during early-stage human infection, as described in one study by where two patients were first treated with antibiotics and subsequently reinfected with their own H. pylori strain [bib_ref] A mutation burst during the acute phase of Helicobacter pylori infection in..., Linz [/bib_ref]. We noted a very high proportion of nonsynonymous mutations among the unique polymorphisms in the reisolates, which are likely to have evolved de novo since first colonization of a new host. Seventy-six percent of the mutations were nonsynonymous (nonsynonymous/synonymous ratio 3.76). This ratio is similar to the ratio of 2.8 observed in the previous study with challenge strain BCM-300 but substantially higher than the ratio of 0.95 observed for four pairs of genomes from sequential H. pylori isolates cultured from chronically infected individuals with a time interval of 3 years [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref]. It has been well documented that nonsynonymous mutations are more common in populations that have undergone a recent strong expansion (as would have happened after experimental transmission). Many slightly deleterious nonsynonymous mutations are lost from the population with time, as shown for H. pylori by the comparisons between short-term (years [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] and long-term (millennia) mutation rates [bib_ref] Microevolution of Helicobacter pylori during prolonged infection of single hosts and within..., Morelli [/bib_ref]. However, the fact that in the reisolates of the present study, nonsynonymous mutations were strongly enriched in genes coding for specific functional classes, such as bacterial envelope traits and host interaction molecules, strongly suggests that the observed patterns of mutations reflect early adaptation to the new host rather than a random distribution of near-neutral mutations.
We recently analyzed patterns of genetic diversity in genomes from H. pylori isolates from different regions of the stomach, using 10 isolates per location. This analysis provided evidence that genes involved in chemotaxis, motility, transport, and cellular processes, as well as OMP-related genes, were particularly prone to diversify in vivo ("high-frequency host variable genes") [bib_ref] Within-host evolution of Helicobacter pylori shaped by niche-specific adaptation, intragastric migrations and..., Ailloud [/bib_ref]. Despite the short time of infection in our present study, several genes that contained unique polymorphisms (oppB, arcS, hcpD, hp0953, and hp0130) overlap the list of high-frequency host variable genes from the analysis by Ailloud et al. [bib_ref] Within-host evolution of Helicobacter pylori shaped by niche-specific adaptation, intragastric migrations and..., Ailloud [/bib_ref] , again suggesting that the patterns of diversity observed reflect in vivo selection rather than random mutational patterns.
Genes with an outer membrane-related role have been shown to vary at higher frequencies than other genes in H. pylori [bib_ref] Bidirectional genomic exchange between Helicobacter pylori strains from a family in Coventry,..., Krebes [/bib_ref] [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref] [bib_ref] In vivo sequence variation in HopZ, a phase-variable outer membrane protein of..., Kennemann [/bib_ref]. In agreement with previous findings, our data emphasize that most of the SNPs identified were again in genes encoding outer membrane proteins. Unique SNPs within babA, the gene encoding the best-characterized outer membrane protein of H. pylori, the Le b binding adhesin, were detected in several reisolates of this study. One reisolate (29C8) harbored a deletion of approximately 1 kb affecting babA. In agreement with this, BabA expression was not detected in this strain. Both frequent mutations and complete loss of BabA expression were reported during infection in primates and humans [bib_ref] Modification of Helicobacter pylori outer membrane protein expression during experimental infection of..., Solnick [/bib_ref] [bib_ref] Expression of the BabA adhesin during experimental infection with Helicobacter pylori, Styer [/bib_ref] [bib_ref] Dynamic expression of the BabA adhesin and its BabB paralog during Helicobacter..., Hansen [/bib_ref]. Frequent mutations in babA and signs of positive (diversifying) selection have been interpreted as signs of adaptation to individual hosts or different niches within one host (e.g., antrum versus corpus). In the case of strain BCS 100, the frequent changes in babA are less easily explained by selection, because clone H1 (or the reisolates) did not exhibit binding to Le b antigen, despite its expression of the adhesin protein BabA. Previously, it was shown that eight amino acids are responsible for the binding to Le b [bib_ref] Structural basis of Lewis(b) antigen binding by the Helicobacter pylori adhesin, Hage [/bib_ref]. A possible reason for the absence of binding by clone H1 to Le b might be a single amino acid substitution in the amino acid sequence of BabA affecting the Fuc4 binding site. Specifically, Asn206, shown to be involved in Le b binding in strain J99, is substituted by a positively charged Glu in clone H1 and in all derivatives of challenge strain BCS 100. This might explain the general lack of Le b binding of all tested clones of the input and output from this experimental infection. While adhesion is considered a pivotal part of H. pylori pathobiology, and BabA is the by-far best-characterized adhesin [bib_ref] Attachment of Helicobacter pylori to human gastric epithelium mediated by blood group..., Borén [/bib_ref] , it has also long been known that not all strains bind to Le b [bib_ref] Metastability of Helicobacter pylori bab adhesin genes and dynamics in Lewis b..., Backstrom [/bib_ref] , and the functional role of the variant version of BabA expressed by H1 is unknown. It has been proposed that BabA may serve a still-uncharacterized role explaining the strong selective pressure against expression of BabA in some model systems [bib_ref] Dynamic expression of the BabA adhesin and its BabB paralog during Helicobacter..., Hansen [/bib_ref]. Our observation that multiple reisolates differed in BabA expression would also be consistent with the hypothesis that BabA might have additional ligands in the absence of Le b binding.
In the same reisolate displaying the partial deletion of babA (29C8), we also observed an intragenomic recombination event between two genes described as essential for the synthesis of type 1 and type 2 Lewis antigens in the bacteria. It was previously suggested that recombination between jhp0562 and the -(1,3) galT gene generates Lewis antigen diversity [bib_ref] Novel functions for the glycosyltransferases Jhp0562 and GalT in Lewis antigen synthesis..., Pohl [/bib_ref]. This mechanism has been suggested to be valuable for H. pylori to mimic the human Lewis antigens on its surface in order to evade the host immune system [bib_ref] Lipopolysaccharide diversity evolving in Helicobacter pylori communities through genetic modifications in fucosyltransferases, Nilsson [/bib_ref] [bib_ref] Simultaneous expression of type 1 and type 2 Lewis blood group antigens..., Monteiro [/bib_ref]. Possibly, strains mimicking the human Le b antigen on their own surface might be prone not to express Le b -binding BabA. This hypothesis needs to be further explored.
One of the striking observations in this study was the unexpected frequency of mutations in genes of the oligopeptide ABC transporter system, suggesting selective evolution during the acute infection of human volunteers. Although the genes were originally annotated as part of the oligopeptide uptake system, an additional role in short peptide uptake has been suggested [bib_ref] Peptide transport in Helicobacter pylori: roles of Dpp and Opp systems and..., Weinberg [/bib_ref]. Both oligopeptide (Opp) and dipeptide (Dpp) transport systems may contribute to the import of carbon and nitrogen sources from the human stomach into the bacterium [bib_ref] Peptide transport in Helicobacter pylori: roles of Dpp and Opp systems and..., Weinberg [/bib_ref] [bib_ref] Helicobacter pylori physiology predicted from genomic comparison of two strains, Doig [/bib_ref]. Several other roles, ranging from the recycling of cell wall peptides to host adhesion, have been attributed to the Opp systems in other microorganisms [bib_ref] Bacterial peptide transporters: messengers of nutrition to virulence, Garai [/bib_ref]. Whether the Opp system in H. pylori similarly has additional functions will have to be addressed in the future, but it seems that the Opp system plays an important role in H. pylori adaptation to a new host, at least in the context of the strain BCS 100. Interestingly, the permease gene oppB was the only gene of the opp operon predicted to be functional in all H. pylori genomes analyzed here, suggesting an important function of this gene either within the Opp system or in a different process. One plausible explanation for the selection of nonfunctional Opp system components in some reisolates could be the absence of specific nutrients from a specific stomach niche. Inactivation of individual Opp system components might help to alleviate metabolic costs of the expression of Opp transporters. Modifications were not found within the opp genes in the reisolates of the infected volunteers challenged with the BCM-300 strain. It was suggested that CagA and VacA might promote colonization in iron-limited environments via facilitating iron acquisition from the host epithelium [bib_ref] Helicobacter pylori perturbs iron trafficking in the epithelium to grow on the..., Tan [/bib_ref]. The authors also proposed that H. pylori must acquire other micronutrients from the host. Thus, it is attractive to speculate that in the absence of bacterial factors like the type 4 secretion system (T4SS) potentially facilitating nutrient acquisition, the bacteria might modulate the uptake of nutrients by altering the activity of permease genes such as the Opp system. However, only 12 reisolates were available in BCM-300, and future studies will be required to test the hypothesis that inflammation status and mutations in opp genes might be linked.
The experimental vaccine used in this trial was a Salmonella strain expressing H. pylori urease. None of the reisolates showed mutations in genes of the urease operon, so there was no evidence for vaccine-induced immune evasion. This differs from our previous study with challenge strain BCM-300 [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref] , where multiple reisolates lost the expression of antigens contained in the experimental vaccine (VacA cytotoxin and CagA), possibly at least in part as a result of vaccine-induced immune selection. In contrast with CagA and VacA, urease expression is essential for H. pylori infection, which is one of the reasons why many vaccine studies have used urease as antigen. The BCS 100 strain is cagPAI Ϫ while the challenge strain BCM-300 used in the other study is cagPAI ϩ . This difference leads very likely to differences in inflammation in the stomach during short-term infection, which might, in the BCS 100 study, reduce the selective pressure by oxidative stress and the immune response.
Another interesting finding was the duplication of the katA gene in the reisolate 103C8. As a result of the duplication of katA that had occurred during infection, this reisolate displayed higher catalase activity and was less sensitive to oxidative stress induced by paraquat. Hence, higher catalase production and activity might be beneficial for H. pylori infecting a host with an increased inflammatory response, since such processes release more reactive oxygen species (ROS). Based on the histology data (28), the volunteer 103 from whom reisolate 103C8 was cultured had one of the highest scores of inflammation (gastroscopy, 6 weeks postinfection [wpi]). In addition, H. pylori was suggested to use its catalase activity to survive ROS caused by phagocytosis [bib_ref] Helicobacter pylori induces but survives the extracellular release of oxygen radicals from..., Ramarao [/bib_ref].
Since the corresponding antrum reisolate 103A4 did not have a second catalase gene, it seems reasonable to consider that within the H. pylori population infecting this volunteer there was a subpopulation of reisolates that were selected, possibly in the corpus of the volunteer, able to resist higher levels of oxidative stress.
Methylation regulates the expression of several genes allowing the microorganisms to respond rapidly to external signals due to changes in DNA methylation patterns. In the present study, using a combination of SMRT sequencing technology and restriction of genomic DNA (gDNA) with commercially available restriction enzymes, we confirmed 24 methylated motifs in the challenge strain and reisolates. Most of the motifs were assigned to already known MTase specificities, but we found methylated sites not described before (CCA m6 AK and GCRC m6 A) or found in other H. pylori strains but not assigned to any MTase (GGCA m6 A and VCGR m6 AG). The inactivation of candidate MTase genes and subsequent SMRT sequencing of the mutants allowed us to discover three novel R-M systems. Differences in the methylomes between challenge strain and reisolates were due to phase variation of two R-M systems methylating CCA m6 AK and CY m6 AN 6 TRG. To our knowledge, this is the second study showing an in vivo switch in the activity of MTase genes in H. pylori during acute or early-stage infection [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref]. This strongly suggests that the reversible ON/OFF switch of these enzymes confers an adaptive advantage of H. pylori in new niches, possibly via the transcriptional regulation of multiple genes. Phase-variable MTases can contribute to bacterial pathogenesis via modifying the expression of multiple genes that together form a "phasevarion" (phasevariable regulon of genes) [bib_ref] The phasevarion: phase variation of type III DNA methyltransferases controls coordinated switching..., Srikhanta [/bib_ref]. Phasevarions were identified in many bacterial pathogens including H. pylori [bib_ref] Haemophilus influenzae phasevarions have evolved from type III DNA restriction systems into..., Fox [/bib_ref] [bib_ref] Phasevarions mediate random switching of gene expression in pathogenic Neisseria, Srikhanta [/bib_ref] [bib_ref] Phasevarion mediated epigenetic gene regulation in Helicobacter pylori, Srikhanta [/bib_ref]. The flaA gene and the outer membrane protein gene hopG were differentially regulated when one yype III MTase, modH5, was not active in H. pylori strain P12 [bib_ref] Phasevarion mediated epigenetic gene regulation in Helicobacter pylori, Srikhanta [/bib_ref]. The novel phase-variable MTase M.HpyH1I that methylates CCAAK is a homolog of ModH5 in the P12 strain (79.37% nucleotide sequence identity), both located downstream of the ATP-dependent DNA helicase RecG. There are multiple CCAAK motifs in recG (both within the coding sequence and immediately upstream of the start codon). Hence, a potential role of CCAAK methylation in the transcriptional control of recG and other genes is possible and will be the subject of future studies. In addition, a role for phasevarions in otitis media diseases has been proposed based on the distribution of phase-variable MTases in Moraxella catarrhalis [bib_ref] ModM DNA methyltransferase methylome analysis reveals a potential role for Moraxella catarrhalis..., Blakeway [/bib_ref]. In pathogenic Neisseria, phasevarions produce different bacterial populations with distinct abilities in colonization [bib_ref] Phasevarions mediate random switching of gene expression in pathogenic Neisseria, Srikhanta [/bib_ref]. Thus, it is conceivable that the modulation of the expression of several genes by the activity of phase-variable MTase genes can contribute to the adaptation of H. pylori to new niches, and this will be explored in future studies.
In conclusion, human volunteer challenge studies performed in the context of vaccine trials are a powerful approach to study short-term adaptation of pathogens to a new host. One unavoidable limitation of this approach is due to the fact that volunteer challenge studies can be performed only in adults while most natural H. pylori infections occur in children. Using one of two such studies available for H. pylori experimental vaccination, we found that the gastric pathogen H. pylori copes with the adaptation requirements during early infection by altering several genes and their functions. This concerns mostly genes and proteins with a role in surface modulation. However, many reisolates after short-term colonization seem to develop quite different and diverse strategies, via modifying diverse genes and parts of their genomes. This diversity implies that the genetic and phenotypic malleability and ways to adapt in H. pylori are manifold and that various adaptation paths in the face of environmental pressures are possible. Among other mechanisms, global coordinated gene regulation via phasevarions might contribute substantially to the early-stage adaptation of H. pylori to new niches.
# Materials and methods
Challenge strain and reisolates. H. pylori strain BCS 100 (ATCC BAA-945) was isolated from an infected individual with mild superficial gastritis and described elsewhere [bib_ref] Challenge model for Helicobacter pylori infection in human volunteers, Graham [/bib_ref]. H. pylori BCS 100 was used to challenge human volunteers as a part of a vaccination study, administering an inoculum dose of 2 ϫ 10 5 bacteria (CFU) [bib_ref] Correlation of T cell response and bacterial clearance in human volunteers challenged..., Aebischer [/bib_ref]. Pairs of reisolates from the antrum and corpus location in the stomach of each infected individual were obtained 10 weeks after the infection. The bacterial samples were isolated from volunteers belonging to control or vaccination groups (see [fig_ref] TABLE 1: Key characteristics of the genome sequences of H [/fig_ref] in the supplemental material). In contrast to classical microbiology technique, input strain BCS 100 was maintained without singlecolony purification, conserving within-strain variation. All genome sequences analyzed in this study were obtained from single-colony purified clones. Clones H1 through H16 were single-colony purified from strain BCS 100 at the Max Planck Institute for Infection Biology at the time when the clinical trial was prepared. Cultures used for this study were inoculated using stock cultures made at this time, so that the number of passages since the isolation of the individual clones was kept low, below five. All genome sequences from reisolates were also obtained from single-colony purified clones. For each of the 10 volunteers studied (five volunteers randomly selected from the vaccine and control groups, respectively), one reisolate from the antrum and one from the corpus were analyzed.
Bacterial strains and culture conditions. H. pylori strains were cultured on blood agar plates (blood agar base II; Oxoid, Wesel, Germany) containing 10% horse blood and supplemented with antibiotics (vancomycin [10 mg/liter], polymyxin B [3.2 mg/liter], amphotericin B [4 mg/liter], and trimethoprim [5 mg/liter]) as previously described [bib_ref] The nucleotide excision repair (NER) system of Helicobacter pylori: role in mutation..., Moccia [/bib_ref]. H. pylori liquid cultures were performed in brain heart infusion broth (BHI; BD Difco, Heidelberg, Germany) with yeast extract (2.5 g/liter), 10% heat-inactivated horse serum, and the same combination of antibiotics as on the blood agar plates [bib_ref] Genome-wide analysis of chromosomal import patterns after natural transformation of Helicobacter pylori, Bubendorfer [/bib_ref]. Escherichia coli strains were grown in LB medium (Lennox L broth; Invitrogen Life Technologies, Darmstadt, Germany) supplemented with ampicillin (200 g/ml) and/or kanamycin (20 g/ml) as described previously [bib_ref] The complex methylome of the human gastric pathogen Helicobacter pylori, Krebes [/bib_ref].
DNA techniques and next-generation sequencing. All procedures were performed following the manufacturer's protocols. Genomic DNA (gDNA) was purified using Genomic-tip 100/G columns (Qiagen, Hilden, Germany). The QIAprep Spin Miniprep kit (Qiagen, Hilden, Germany) was utilized for E. coli plasmid isolation.
Complete genome sequencing of clone H1 from the challenge strain BCS 100 and the reisolates was performed using SMRT sequencing technology on a Pacific Biosciences RSII instrument. Preparation of SMRTbell template libraries was carried out as described previously [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref]. One SMRT cell was sequenced per strain using P6/C4 chemistry. De novo genome assemblies were carried out using Hierarchical Genome Assembly Process 3 ("RS_HGAP_Assembly.3" protocol) within SMRT Portal 2.3.0. Complete genomes have been circularized and rotated to the dnaA gene as starting position. The standardized "RS_Modification_and_Motif_Analysis.1" protocol was used applying default parameters to detect methylated bases and to identify the motifs as performed in references 20 and 41.
Genome sequences of the strains H2 to H16 were obtained using an Illumina MiSeq system as described before [bib_ref] Genome-wide analysis of chromosomal import patterns after natural transformation of Helicobacter pylori, Bubendorfer [/bib_ref]. Libraries were prepared using the Nextera DNA sample preparation kit (Illumina, San Diego, CA, USA). The length of the fragments was calculated applying the high-sensitivity DNA analysis kits (Agilent Technologies, Palo Alto, CA, USA) in an Agilent 4200 TapeStation device. Illumina MiSeq 2 ϫ 300 cycle reagent kit v3 was employed to sequence the libraries in a MiSeq sequencer. De novo assembly of the paired-end reads was performed using SPAdes 3.9.0 with default parameters [bib_ref] SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing, Bankevich [/bib_ref].
Genome comparison. The genome of clone H1 was annotated using the Rapid Annotation Server (RAST). KODON (Applied Maths, Sint-Martens-Latem, Belgium) and Geneious 11.0.2 (69) were used for whole-genome comparison of H1 (used as reference) with the reisolates and clones H2 to H16. As described previously, differences were classified as single nucleotide polymorphisms (SNPs) or as clusters of polymorphisms (CNPs). CNPs consist of a group of polymorphisms within less than 200 contiguous bp that are flanked by at least 200 bp of identical sequence on both sides. CNPs are, from previous evidence, most likely the result of allelic replacement events after homoeologous recombination [bib_ref] Recombination and mutation during long-term gastric colonization by Helicobacter pylori: estimates of..., Falush [/bib_ref] [bib_ref] Helicobacter pylori genome evolution during human infection, Kennemann [/bib_ref]. Phylogenetic inference was performed with M r Bayes 3.2 (70), using 1,000,000 generations and 10% burn-in.
Selection and inactivation of candidate MTase genes and graphical representation. Identification of R-M genes within the genome of reference clone H1 was performed as described in reference 71. Using the REBASE database (42), we predicted the specificities of previously described MTase genes and selected candidate genes encoding putative MTases. The candidate genes were inactivated via insertion of an aphA3 cassette conferring resistance to kanamycin as described previously [bib_ref] The complex methylome of the human gastric pathogen Helicobacter pylori, Krebes [/bib_ref]. Plasmids containing the interrupted gene were used for natural transformations using H. pylori clone H1 or the reisolate 12C8 as recipient. Successful allelic exchange between the plasmid and the chromosomal target gene was confirmed by PCR. All primers and strains used are listed in [fig_ref] TABLE 2: Unique and pair-specific SNPs in H [/fig_ref] and B, respectively.
Graphical representation of motifs within the H1 genome was generated using Circos (72) [fig_ref] FIG 4: Circos plot displaying the distribution of methylated sequence motifs in H [/fig_ref]. Restriction assays. Isolated H. pylori gDNA (250 to 300 ng) was used in a 20-l reaction mixture containing the corresponding amount of restriction enzyme buffer (NEB, Frankfurt am Main, Germany), 1 l of the suitable restriction enzyme (NEB, Frankfurt am Main, Germany), and high-pressure liquid chromatography (HPLC) water. A 20-l sample with the same amount of gDNA, restriction enzyme buffer, and HPLC water without enzyme was used as negative control. Both tubes were incubated at 37°C for 1 h. Then, 10 l of the reactions was loaded onto a 1% agarose gel to detect the digested/undigested gDNA patterns.
Isolation and detection of proteins. H. pylori from 22-to 24-h blood agar plate cultures was harvested in 1 ml cold phosphate-buffered saline (PBS) (600 ϫ g, 4°C, 5 min). Cell pellets were suspended in 200 l of homogenization buffer (Tris-HCl, 100 mM, pH 7.4). Samples were sonicated for 1 min at continuous pulse. Total protein concentration was calculated using the Pierce bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific, Darmstadt, Germany). The same amounts of protein samples were separated on SDS-PAGE gels (14%). Subsequently, Western blotting (WB) was performed for the detection of target proteins. Detection of BabA was achieved with a rabbit antiserum kindly provided by Thomas Borén (Umeå University, Sweden) and a secondary peroxidase-labeled Affini-pure goat antirabbit IgG (HϩL) antibody (Jackson Immuno Research Laboratories, USA). Catalase detection was achieved using an antibody from the Ridascreen FemtoLab stool antigen test (R-Biopharm AG, Darmstadt, Germany) as described previously [bib_ref] Localisation and protein-protein interactions of the Helicobacter pylori taxis sensor TlpD and..., Behrens [/bib_ref].
SuperSignal West Pico chemiluminescent substrate (Thermo Fisher Scientific, Darmstadt, Germany) was used for WB developing. The Las-3000 imaging system (Fujifilm Life Science, Düsseldorf, Germany) and chemiluminescence were used to reveal the WB.
qPCR. Quantitative PCR (qPCR) was performed as described before [bib_ref] The core genome m5 C methyltransferase JHP1050 (M.Hpy99III) plays an important role..., Estibariz [/bib_ref]. Synthesis of cDNA was performed with SuperScript III reverse transcriptase (Thermo Fisher Scientific, Darmstadt, Germany) using 1 g of RNA. A Bio-Rad CFX96 system was used to perform the qPCR with specific primers [fig_ref] TABLE 2: Unique and pair-specific SNPs in H [/fig_ref] and SYBR green Master Mix (Qiagen, Hilden, Germany). Samples were normalized to an internal 16S rRNA control. Reaction conditions in agreement with the MIQE (minimum information for publication of quantitative real-time PCR experiments) guidelines are available in the supplemental material. GraphPad Prism 7 (GraphPad Software, La Jolla, CA, USA) was used to compile all graphs.
Oxidative stress assay. The oxidative stress assay was carried out as described previously [bib_ref] Role of energy sensor TlpD of Helicobacter pylori in gerbil colonization and..., Behrens [/bib_ref]. Liquid cultures were inoculated with challenge strain clone H1 or with the reisolate 103C8 and grown overnight (37°C, 175 rpm, microaerobic conditions, initial optical density at 600 nm [OD 600 ] ϭ 0.06). Afterward, 10 M paraquat (PQ) (paraquat dichloride hydrate [Pestanal], number 36541; Sigma-Aldrich, Germany) was added to the cultures; controls were left untreated. The time point of the inoculation was counted as time zero. Serial dilutions were plated onto blood agar plates at time zero and after 10 h. The plates were incubated at 37°C and under microaerobic conditions. Finally, after 4 to 5 days of incubation, the number of colonies was counted. Data were normalized to time point 0.
Catalase activity. Catalase activity was determined using the Megazyme catalase assay kit (Megazyme, Butzbach, Germany) following the manufacturer's instructions. Bacteria from 20 to 22-h blood agar plate cultures were harvested and suspended in ice-cold 1ϫ PBS to an OD 600 of 1. Lysates were obtained as mentioned above, and 1:1,000 dilutions of protein lysates were used to perform the assays. Three independent biological replicates were carried out for each sample.
Le b binding assay. The Le b binding assay was carried out as described previously [bib_ref] Genome and methylome variation in Helicobacter pylori with a cag pathogenicity island..., Nell [/bib_ref] , with minor modifications. Bacteria from blood agar plates (incubated for 22 to 24 h) were suspended in 1 ml PBS and centrifuged (2,800 ϫ g, 5 min, 4°C). The OD 600 was measured and adjusted to 2 ϫ 10 8 cells in 450 l of PBS for subsequent biotinylation during 1 h with N-hydroxysuccinimide-long chain (NHS-LC) biotin (succinimidyl-6-(biotinamido)hexanoate, Thermo Fisher Scientific, Darmstadt, Germany) (125 g/25 l). Then, 250 ng of bovine serum albumin (BSA) or 250 ng of Le b -BSA was used to coat a 96-well covalent microtiter plate (Corning Costar, USA). Immobilization of the glycoproteins was performed under UV light for 30 s, and afterward, the plate was blocked with PBS containing 5% BSA. Subsequently, 50 l per well of biotinylated bacteria was coincubated in the dark for 1 h and washed 3 times with PBS. Sample fixation was achieved with 100 l of 2% paraformaldehyde (100 mM potassium phosphate buffer, neutral pH). The plate was washed 3 times with wash buffer (0.05% Tween 20 in PBS), blocked for 1 h with assay diluent (PBS containing 10% fetal calf serum), and again washed 5 times with wash buffer. After an incubation step of 90 min at room temperature with neutravidin-horseradish peroxidase (HRP) conjugate in assay diluent, the plate was rinsed again and incubated with tetramethylbenzidine (TMB) substrate (BD Biosciences). Using 50 l per well of 1 M H 3 PO 4 , the reaction was stopped. The signal was detected in a microplate reader at 450 nm. Only one replicate per sample (in technical triplicates) was performed since no binding was detected for the strain H1 while a clear binding was determined for the positive control, strain J99.
Data availability. Sequence data have been deposited in the NCBI with link to BioProject accession PRJNA522954 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA522954).
# Supplemental material
Supplemental material is available online only.
[fig] FIG 1: A) Schematic representation of the vaccine/challenge trial [/fig]
[fig] FIG 2: In vivo rearrangement in the reisolate 103C8 due to gene fusions and duplication of the katA gene. (A) Representation of the genomic context of katA in H1 and the duplication of katA and a fusion of duplicated fragments of the genes frpB and kapA in reisolate 103C8. (B) [/fig]
[fig] FIG 3: Graphical representation of the H. pylori oligopeptide transport (opp) gene cluster and genetic changes observed in BCS 100 and reisolates from infected volunteers. (A) Schematic representation of the opp gene clusters in E. coli and H. pylori. opp genes form a contiguous gene cluster in E. coli, while the four opp genes in H. pylori are located in two different loci. H. pylori does not have a homolog of oppF. (B) Graphical representation of the opp gene configurations found in strainswhose genomes were sequenced in this study. Based on the nucleotide sequences, genes in blue are predicted to be truncated while genes in gray are likely to be active. Note that all reisolates and challenge strain clones H2 to H16 are predicted to express all four opp genes.Estibariz et al. ® to the Lewis b (Le b ) histo-blood group antigen(37,38) [/fig]
[fig] FIG 4: Circos plot displaying the distribution of methylated sequence motifs in H. pylori BCS 100 clone H1. Every circle of colored tracks represents one MTase target motif; the legend at the right side of the figure states the order of motifs represented by the 22 rings, from outer to inner circle. The motif CAAK, which is methylated in some reisolates, is not depicted, because it is not methylated in clone H1.Estibariz et al. [/fig]
[fig] FIG S1 ,: PDF file, 0.5 MB. FIG S2, PDF file, 0.6 MB. FIG S3, PDF file, 0.5 MB. FIG S4, PDF file, 0.4 MB. [/fig]
[table] TABLE 1: Key characteristics of the genome sequences of H. pylori reisolates from human volunteers challenged with strain BCS 100 a [/table]
[table] TABLE 2: Unique and pair-specific SNPs in H. pylori reisolates from infected volunteers a [/table]
[table] TABLE 3: Methylated sequence motifs detected by SMRT sequencing in clone H1 of challenge strain BCS 100 and reisolates from human volunteers a [/table]
[table] TABLE S1 ,: PDF file, 0.5 MB. TABLE S2, PDF file, 0.1 MB. TABLE S3, PDF file, 0.4 MB. TABLE S4, PDF file, 0.6 MB. TABLE S5, PDF file, 0.5 MB. TABLE S6, PDF file, 0.5 MB. [/table]
|
Treatment-Related Adverse Events with PD-1 or PD-L1 Inhibitors: A Systematic Review and Meta-Analysis
Objective: to evaluate the risk of treatment-related adverse events of different severity and different system with PD-1 or PD-L1 inhibitors. Methods: randomized controlled trials (RCTs) that using PD-1/PD-L1 for cancer treatment were searched in the PubMed, Embase, Cochrane Library, and Web of Science from 1 January 2019 to 31 May 2021. Adverse events data were extracted from clinical trials website or original article by two authors separately. Meta-analysis was used to determine risk ratio (RR) and 95% confidence interval (95% CI) of adverse events in PD-1/PD-L1 inhibitors groups compared to that of control groups. Subgroup analyses were also performed. Results: a total of 5807 studies were initially identified and after exclusion, 41 studies were included in meta-analysis. All the trials were international multicenter, randomized, phase II/III clinical trials, with the median follow-up of 27.5 months on average. Analysis of all grade adverse events showed that PD-1/PD-L1 inhibitors treatment significantly increased the risk of immune-related adverse events, including pruritus (RR: 2.34, 95% CI: 1.85-2.96), rash (RR: 1.53, 95% CI: 1.25-1.87), ALT elevation (RR 1.54, 95% CI 1.23-1.92), AST elevation (AST: RR 1.49, 95% CI 1.20-1.85), hepatitis (RR: 3.54, 95% CI: 1.96-6.38) and hypothyroid (RR: 5.29, 95% CI: 4.00-6.99) compared with that of control group. Besides that, PD-1/PD-L1 inhibitors were associated with higher risk of adverse events related to respiratory system including cough (RR: 1.33, 95% CI: 1.21-1.48), dyspnea (RR:1.23, 95% CI: 1.12-1.35) and chest pain (RR: 1.26, 95% CI: 1.07-1.47) compared with that of control groups in our meta-analysis and the dyspnea was taken high risk both in all grade and grade 3 or higher (RR: 1.55, 95% CI: 1.13-2.12). The risk of arthralgia was increased with PD-1/PD-L1 inhibitors (RR: 1.27, 95% CI: 1.10-1.47). Although the risk of myalgia was similar with PD-1/PD-L1 inhibitors and control groups, under subgroup analysis, PD-1/PD-L1 inhibitors decreased the risk of myalgia (RR: 0.56, 95% CI: 0.45-0.70) compared with that of chemotherapy. Conclusions: our results provide clear evidence that the risk of treatment-related adverse events in PD-1 or PD-L1 varies widely in different system. In particular, when using PD-1/PD-L1 inhibitors for oncology treatment, besides the common immune-related adverse events like pruritus, rash, hepatitis, and hypothyroid, the respiratory disorders and musculoskeletal disorders, such as cough, dyspnea, arthralgia, and myalgia, should also be taken into consideration.
# Introduction
In recent years, cancer patients gained increasingly significant benefits from immunotherapy, due to its remarkable clinical efficacy and durable response [bib_ref] PD-1 and PD-L1 in cancer immunotherapy: Clinical implications and future considerations, Jiang [/bib_ref]. Up to now, the FDA approved PD-1 inhibitors nivolumab, pembrolizumab, and PD-L1 inhibitors atezolizumab, avelumab, and durvalumab in clinical trials [bib_ref] Systematic Review of PD-1/PD-L1 Inhibitors in Oncology: From Personalized Medicine to Public..., Chang [/bib_ref]. In China, a number of PD-1 or PD-L1 inhibitors successfully entered the Medicare list, such as sintilimab, tislelizumab, and camrelizumab for the treatment of metastatic colorectal cancer, nonsmall cell lung cancer, Hodgkin's lymphoma, and metastatic melanoma.
The PD-1/PD-L1 axis can be modulated by various signals in cancer cells because of higher expression of PD-1/PD-L1 in a large number of solid cancers, which plays a critical role in oncology treatment [bib_ref] PD-1/PD-L1 pathway: Current researches in cancer, Han [/bib_ref]. In nonsmall-cell lung cancer, immune checkpoint inhibitors show very robust efficacy over docetaxel in overall survival, whether combined with chemotherapy or not [bib_ref] Comparative efficacy and safety of immunotherapies targeting the PD-1/PD-L1 pathway for previously..., Almutairi [/bib_ref] [bib_ref] Anti-PD-1 versus anti-PD-L1 therapy in patients with pretreated advanced non-small-cell lung cancer:..., Tartarone [/bib_ref] [bib_ref] A pooled meta-analysis of PD-1/L1 inhibitors incorporation therapy for advanced non-small cell..., Wan [/bib_ref]. The combination of PD-1/PD-L1 immune checkpoint inhibitors with first-line chemotherapy provide a significant benefit for extensive-stage small cell lung cancer in overall survival, progression-free survival and objective response rate [bib_ref] Adding PD-1/PD-L1 Inhibitors to Chemotherapy for the First-Line Treatment of Extensive Stage..., Facchinetti [/bib_ref]. For the treatment of advanced renal cell carcinoma and gastroesophageal cancer, PD-1/PD-L1 inhibitors were proved to improve antitumor activity and overall survival [bib_ref] Efficacy and safety of anti-PD-1/PD-L1 agents vs chemotherapy in patients with gastric..., Wang [/bib_ref] [bib_ref] Outcomes Associated with First-Line anti-PD-1/PD-L1 agents vs. Sunitinib in Patients with Sarcomatoid..., Buonerba [/bib_ref] [bib_ref] Comparative Efficacy and Safety of Programmed Death-1 Pathway Inhibitors in Advanced Gastroesophageal..., Da Silva [/bib_ref].
Although PD-1/PD-L1 drugs outperformed in prolonging patients' survival, research on drug toxicity and adverse events is limited. With the increase use immune checkpoint inhibitors, there were more and more reports on immune-related adverse events [bib_ref] Immune-related adverse events of checkpoint inhibitors, Ramos-Casals [/bib_ref]. In a meta-analysis of single-arm PD-1/PD-L1 inhibitors, Wang et al. found that fatigue was the most common all-grade treatment-related adverse events, followed by pruritus and diarrhea [bib_ref] Treatment-Related Adverse Events of PD-1 and PD-L1 Inhibitors in Clinical Trials: A..., Wang [/bib_ref]. A Cochrane systematic review concluded that the frequency of Grade 3-4 adverse events of single immune checkpoint inhibitor was low, but they did not report specific adverse events [bib_ref] Single or combined immune checkpoint inhibitors compared to first-line platinum-based chemotherapy with..., Ferrara [/bib_ref]. pointed out that musculoskeletal problems may be a common adverse event with anti-PD-1 drugs, while further investigation was required [bib_ref] Immune-related adverse events for anti-PD-1 and anti-PD-L1 drugs: Systematic review and meta-analysis, Baxi [/bib_ref]. Although different PD-1 and PD-L1 inhibitors are accompanied by various treatment-related adverse events, there seems to be no difference between them [bib_ref] Treatment-Related Adverse Events of PD-1 and PD-L1 Inhibitors in Clinical Trials: A..., Wang [/bib_ref] [bib_ref] Use of Immunotherapy With Programmed Cell Death 1 vs Programmed Cell Death..., Duan [/bib_ref].
Treatment-related adverse events refer to an unfavorable change in the health of a participant in clinical trials, which is an important issue in oncology [bib_ref] Toxicities of the anti-PD-1 and anti-PD-L1 immune checkpoint antibodies, Naidoo [/bib_ref]. A comprehensive understanding of adverse events with PD-1 and PD-L1 inhibitors is essential to clinical practice. In this study, we conducted a systematic review and meta-analysis to evaluate the relative risk of treatment-related adverse events with PD-1 and PD-L1 inhibitors in different severity and different system compared with that of control group. We also performed subgroup analysis to investigate the safety of PD-1/PD-L1 inhibitors.
# Methods
## Search strategy
We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A comprehensive literature search was performed, independently by two authors (Z.Y. and La B.), under four databases, PubMed, Embase, Cochrane Library, and Web of Science, from 1 January 2019 to 31 May 2021. Taking PubMed as an example, the following search terms were used for study retrieval: ("Immune Checkpoint Inhibitors" OR "immune checkpoint inhibitor"[tiab] OR *checkpoint inhibitor*[tiab] OR checkpoint block* OR "PD 1"[tiab] OR "PD L1"[tiab] OR "anti PD-1"[tiab] OR "anti-PD-L1" . We did not impose any restrictions on type of article, language, design method, cancer type, or number of populations at risk. A similar search strategy and search terms were repeated in Embase, Cochrane Library, and Web of Science, respectively. In addition, reference lists of potentially relevant reports and reviews were screened to identify other eligible studies. We combined the search results in a bibliographic management software (EndNote X9).
## Study selection
Three reviewers (Z.Y., L.B., and G.Y.) independently identified articles eligible for in-depth examination using the following predefined inclusion and exclusion criteria: (1) study design: randomized controlled clinical trials with the purpose of cancer therapy were considered. We excluded case reports, case series, single-arm cohort studies, reviews and meeting abstracts, but we had no restriction on cancer type; (2) type of interventions: participants were treated with a single-agent PD-1 or PD-L1 inhibitor in treatment group;
(3) type of outcomes: we focused on treatment-related adverse events, so the included studies should display reported tabulated data on treatment-related adverse events in clinicaltrail.gov or in the full article; (4) published in English. Two authors (Z.Y. and La B.) screened all titles and abstracts for full text reviews. Disagreements were resolved by consensus involving three authors (Z.Y., L.B., and G.Y.).
## Data extraction
Data from each study were extracted by two authors (Z.Y. and La B.). Differences were resolved by consensus. The trial name, phase, cancer type, PD-1 and PD-L1 inhibitor used, dose escalation, dose schedule, number of patients, median follow-up, and number of all treatment-related adverse events were obtained from each included study. Treatment-related adverse events that we care about included general disorders (fatigue, fever, headache), skin disorders (pruritus, rash), respiratory disorders (cough, dyspnea, chest pain, pneumonia), gastrointestinal disorders (loss of appetite, nausea, vomiting, diarrhea, constipation, abdominal pain), liver disorders (ALT elevation, AST elevation, hepatitis), endocrine disorders (hypothyroid), musculoskeletal disorders (myalgia, arthralgia), blood disorders (anemia, neutrophil decrease). Serious adverse events (grade 3 or higher) and other adverse events (grade 1-2) were both extracted.
## Outcomes
Our primary outcome was the incidence of different type of treatment-related adverse events. We recorded data on adverse events reported as serious or other on clinical trials website. For data extracted from published reports, we identified grade 3 or higher as serious and grade 1-2 as other. If the study did not report a specific adverse event, we assumed that the event did not occur.
## Risk of bias assessment
Two authors (Z.Y. and La B.) evaluated the risk of bias with regard to adverse event outcomes by using the tool recommended by the Cochrane Collaboration Handbook. The bias assessment was divided into seven aspects: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other bias, which were presented in a graph with high risk (red color), green color (low-risk), and yellow color (unclear).
## Data synthesis and analysis
For each included study, we calculated risk ratio and 95% confidence interval (95% CI) for incidence rate in the intervention arm compared with that of control, based on the reported number of events and sample size. We used the I 2 index to examine heterogeneity across trials for each outcome. If significant heterogeneity was not present (p > 0.1), pooled risk ratio and 95% CI were estimated with a fixed effect model using the inverse variance method. A random effect model using the inverse variance was used to calculate pooled risk ratio and 95% CI if significant heterogeneity was presents (p < 0.1). If a study included more than one intervention arm, we separately compared each intervention arm with the control arm. For example, Roy S. Herbst reported 2 mg/kg and 10 mg/kg for pembrolizumab [bib_ref] Long-Term Outcomes and Retreatment Among Patients With Previously Treated, Programmed Death-Ligand 1-Positive,..., Herbst [/bib_ref] , and Caroline Robert reported q2w and q3w for pembrolizumab [bib_ref] Pembrolizumab versus ipilimumab in advanced melanoma (KEYNOTE-006): Post-hoc 5-year results from an..., Robert [/bib_ref]. We conducted subgroup analysis by different immune checkpoint inhibitors, different treatment frequency, and different control groups. In subgroup analysis of different control groups, we delivered control groups into chemotherapy, nonchemotherapy, ipilimumab and placebo four subgroups. The trials using "chemotherapy" or "docetaxel" terms as their control were considered in chemotherapy subgroup. "Non-chemotherapy" subgroup included bevacizumab, brentuximab vedotin, dacarbazine, and everolimus as their control groups. We applied the Cochrane Collaboration's tool to assess the risk of bias. Statistical analyses were conducted using 'meta' package in R-4.0.3. The evaluation of bias was based on Review Manager 5.4. shows that the initial research yielded 5,807 relevant articles. After screening and eligibility assessment, we finally included 41 studies in the meta-analysis, a total of 13,232 patients in treatment group and 11,670 in control group. To facilitate the calculation, for the studies with controls of two or more treatments, we divided them into two treatmentcontrol groups [bib_ref] Long-Term Outcomes and Retreatment Among Patients With Previously Treated, Programmed Death-Ligand 1-Positive,..., Herbst [/bib_ref] [bib_ref] Pembrolizumab versus ipilimumab in advanced melanoma (KEYNOTE-006): Post-hoc 5-year results from an..., Robert [/bib_ref]. Finally, we included 45 clinical trials in the meta-analysis. All trials were international multicenter, randomized, phase II/III clinical trials with the median follow-up of 27.5 months on average. The PD-1 and PD-L1 inhibitors used included nivolumab (n = 14), pembrolizumab (n = 16), atezolizumab (n = 6), avelumab (n = 3), durvalumab (n = 2), camrelizumab (n = 1) and cemiplimab (n = 1). The trials involved the treatment of gastrointestinal cancer (n = 8), head and neck squamous cell carcinoma (n = 4), melanoma (n = 6), lung cancer (n = 16), urothelial carcinoma (n = 5), and other cancer (n = 4). Detailed information is summarized in [fig_ref] Table 1: Characteristics of included studies [/fig_ref] pooled risk ratio and 95% CI if significant heterogeneity was presents (p < 0.1). If a study included more than one intervention arm, we separately compared each intervention arm with the control arm. For example, Roy S. Herbst reported 2 mg/kg and 10 mg/kg for pembrolizumab [bib_ref] Long-Term Outcomes and Retreatment Among Patients With Previously Treated, Programmed Death-Ligand 1-Positive,..., Herbst [/bib_ref] , and Caroline Robert reported q2w and q3w for pembrolizumab [bib_ref] Pembrolizumab versus ipilimumab in advanced melanoma (KEYNOTE-006): Post-hoc 5-year results from an..., Robert [/bib_ref]. We conducted subgroup analysis by different immune checkpoint inhibitors, different treatment frequency, and different control groups. In subgroup analysis of different control groups, we delivered control groups into chemotherapy, nonchemotherapy, ipilimumab and placebo four subgroups. The trials using "chemotherapy" or "docetaxel" terms as their control were considered in chemotherapy subgroup. "Non-chemotherapy" subgroup included bevacizumab, brentuximab vedotin, dacarbazine, and everolimus as their control groups. We applied the Cochrane Collaboration's tool to assess the risk of bias. Statistical analyses were conducted using 'meta' package in R-4.0.3. The evaluation of bias was based on Review Manager 5.4. shows that the initial research yielded 5,807 relevant articles. After screening and eligibility assessment, we finally included 41 studies in the meta-analysis, a total of 13,232 patients in treatment group and 11,670 in control group. To facilitate the calculation, for the studies with controls of two or more treatments, we divided them into two treatment-control groups [bib_ref] Long-Term Outcomes and Retreatment Among Patients With Previously Treated, Programmed Death-Ligand 1-Positive,..., Herbst [/bib_ref] [bib_ref] Pembrolizumab versus ipilimumab in advanced melanoma (KEYNOTE-006): Post-hoc 5-year results from an..., Robert [/bib_ref]. Finally, we included 45 clinical trials in the meta-analysis. All trials were international multicenter, randomized, phase II/III clinical trials with the median follow-up of 27.5 months on average. The PD-1 and PD-L1 inhibitors used included nivolumab (n = 14), pembrolizumab (n = 16), atezolizumab (n = 6), avelumab (n = 3), durvalumab (n = 2), camrelizumab (n = 1) and cemiplimab (n = 1). The trials involved the treatment of gastrointestinal cancer (n = 8), head and neck squamous cell carcinoma (n = 4), melanoma (n = 6), lung cancer (n = 16), urothelial carcinoma (n = 5), and other cancer (n = 4). Detailed information is summarized in [fig_ref] Table 1: Characteristics of included studies [/fig_ref]. [fig_ref] Figure 2: Risk of bias graph [/fig_ref] shows the risk of bias evaluated by two authors and summaries the bias of every included trial. All RCTs claimed randomization when grouping. High performance bias and detection bias were considered as seven trials explicitly stated doubleblind when grouping and the rest were all open-label. We considered high reporting bias because the primary outcome of many RCTs is to evaluate the treatment effect, such as overall survival (OS) or progression-free survival (PFS), rather than adverse events.
# Results
## Results of the search
# Results
## Results of the search
## Quality assessment
Life 2021, 11, x FOR PEER REVIEW 8 of 20 [fig_ref] Figure 2: Risk of bias graph [/fig_ref] shows the risk of bias evaluated by two authors and summaries the bias of every included trial. All RCTs claimed randomization when grouping. High performance bias and detection bias were considered as seven trials explicitly stated double-blind when grouping and the rest were all open-label. We considered high reporting bias because the primary outcome of many RCTs is to evaluate the treatment effect, such as overall survival (OS) or progression-free survival (PFS), rather than adverse events. [fig_ref] Table 2: Results of all grades' adverse events [/fig_ref] summarizes the results of all grades' adverse events with overall risk ratio, 95% confidence intervals (95% CI), and assessment of heterogeneity. For general disorders, fatigue was slightly reduced by PD-1 or PD-L1 inhibitor treatment (RR: 0.91, 95% CI: 0.85-0.99) and [fig_ref] Figure 3: Forest plot of all grade fatigue for PD-1/PD-L1 inhibitors compared with that... [/fig_ref] demonstrates the corresponding forest plot. It is interesting that PD-1 or PD-L1 inhibitors showed lower risk of peripheral neuropathy compared with control groups (RR: 0.23, 95% CI: 0.16-0.33). For musculoskeletal disorders, arthralgia is another remarkable finding that PD-1 or PD-L1 inhibitors had a higher risk compared with control groups (RR: 1.27, 95% CI: 1.10-1.47) and the forest plot was shown in [fig_ref] Figure 4: Forest plot of all grade arthralgia for PD-1/PD-L1 inhibitors compared with that... [/fig_ref]. PD-1 or PD-L1 treatment had a great impact on skin disorders. The risk of pruritus was 2.34 (95% CI: 1.85-2.96) times in treatment compared with that of control group, as is shown in [fig_ref] Figure 2: Risk of bias graph [/fig_ref] , and 1.53 (95% CI: 1.25-1.87) times for rash. Regarding the respiratory disorders, PD-1/PD-L1 inhibitors took a higher risk of respiratory disorders, especially for cough (RR: 1.33, 95% CI: 1.21-1.48), dyspnea (RR: 1.23, 95% CI: 1.12-1.35) and chest pain (RR: 1.26, 95% CI: 1.07-1.47). Although the incidence of pneumonia-related symptoms was higher in PD-1 or PD-L1 treatment group, the difference in pneumonia (RR: 0.96, 95% CI: 0.79-1.18) between two groups was not significant, as is shown in [fig_ref] Figure 3: Forest plot of all grade fatigue for PD-1/PD-L1 inhibitors compared with that... [/fig_ref]. As to gastrointestinal disorders, PD-1 or PD-L1 inhibitors were less harmful to gastrointestinal system. The results performed in [fig_ref] Figure 4: Forest plot of all grade arthralgia for PD-1/PD-L1 inhibitors compared with that... [/fig_ref] confirmed that the incidence of nausea was significantly reduced by PD-1/PD-L1 inhibitors (RR: 0.67, 95% CI: 0.57-0.79). The incidence of vomiting decreased one fifth compared with that of control group (RR: 0.79, 95% CI: 0.68-0.92). For liver disorders, PD-1/PD-L1 inhibitors led to apparent liver system damage. [fig_ref] Figure 5: Subgroup analysis of AST elevation in different immune checkpoint inhibitors [/fig_ref] and [fig_ref] Figure 6: Forest plot of subgroup analysis of myalgia in different control groups [/fig_ref] shows that the risk of ALT elevation (RR: 1.58, 95% CI: 1.26-1.99) and AST elevation (RR: 1.56, 95% CI: 1.22-2.00) were significantly increased. What's more, PD-1/PD-L1 inhibitors were associated with higher risk of hepatitis (RR: 3.54, 95% CI: 1.96-6.38), as is shown in . Finally, as to endocrine disorders, the [fig_ref] Table 2: Results of all grades' adverse events [/fig_ref] summarizes the results of all grades' adverse events with overall risk ratio, 95% confidence intervals (95% CI), and assessment of heterogeneity. For general disorders, fatigue was slightly reduced by PD-1 or PD-L1 inhibitor treatment (RR: 0.91, 95% CI: 0.85-0.99) and [fig_ref] Figure 3: Forest plot of all grade fatigue for PD-1/PD-L1 inhibitors compared with that... [/fig_ref] demonstrates the corresponding forest plot. It is interesting that PD-1 or PD-L1 inhibitors showed lower risk of peripheral neuropathy compared with control groups (RR: 0.23, 95% CI: 0.16-0.33). For musculoskeletal disorders, arthralgia is another remarkable finding that PD-1 or PD-L1 inhibitors had a higher risk compared with control groups (RR: 1.27, 95% CI: 1.10-1.47) and the forest plot was shown in [fig_ref] Figure 4: Forest plot of all grade arthralgia for PD-1/PD-L1 inhibitors compared with that... [/fig_ref]. PD-1 or PD-L1 treatment had a great impact on skin disorders. The risk of pruritus was 2.34 (95% CI: 1.85-2.96) times in treatment compared with that of control group, as is shown in [fig_ref] Figure 2: Risk of bias graph [/fig_ref] , and 1.53 (95% CI: 1.25-1.87) times for rash. Regarding the respiratory disorders, PD-1/PD-L1 inhibitors took a higher risk of respiratory disorders, especially for cough (RR: 1.33, 95% CI: 1.21-1.48), dyspnea (RR: 1.23, 95% CI: 1.12-1.35) and chest pain (RR: 1.26, 95% CI: 1.07-1.47). Although the incidence of pneumonia-related symptoms was higher in PD-1 or PD-L1 treatment group, the difference in pneumonia (RR: 0.96, 95% CI: 0.79-1.18) between two groups was not significant, as is shown in [fig_ref] Figure 3: Forest plot of all grade fatigue for PD-1/PD-L1 inhibitors compared with that... [/fig_ref]. As to gastrointestinal disorders, PD-1 or PD-L1 inhibitors were less harmful to gastrointestinal system. The results performed in [fig_ref] Figure 4: Forest plot of all grade arthralgia for PD-1/PD-L1 inhibitors compared with that... [/fig_ref] confirmed that the incidence of nausea was significantly reduced by PD-1/PD-L1 inhibitors (RR: 0.67, 95% CI: 0.57-0.79). The incidence of vomiting decreased one fifth compared with that of control group (RR: 0.79, 95% CI: 0.68-0.92). For liver disorders, PD-1/PD-L1 inhibitors led to apparent liver system damage. [fig_ref] Figure 5: Subgroup analysis of AST elevation in different immune checkpoint inhibitors [/fig_ref] shows that the risk of ALT elevation (RR: 1.58, 95% CI: 1.26-1.99) and AST elevation (RR: 1.56, 95% CI: 1.22-2.00) were significantly increased. What's more, PD-1/PD-L1 inhibitors were associated with higher risk of hepatitis (RR: 3.54, 95% CI: 1.96-6.38), as is shown in . Finally, as to endocrine disorders, the adverse event of treatment on endocrine is mainly thyroid dysfunction. shows that the hypothyroid of using PD-1 or PD-L1 was 5.29 (95% CI: 4.00-6.99) times compared with that of control group.
## Quality assessment
## Results of treatment-related adverse event
## All grade adverse events
## Results of treatment-related adverse event
## All grade adverse events
## Grade 3 or higher adverse event
The incidence of adverse events of grade 3 or higher was low-level both in the treatment group and the control group. [fig_ref] Table 3: Results of Grade 3 or higher adverse events [/fig_ref] summarizes the number of trials, corresponding risk ratio and heterogeneity analysis of grade 3 or higher adverse events using PD-1/PD-L1 inhibitors. It is apparent from this table that PD-1/PD-L1 inhibitors took highly risk in serious dyspnea compared with that of control groups (RR: 1.55, 95% CI: 1.13-2.12). No statistically significant difference was found in the incidence of serious general adverse events (fatigue: RR = 0.78, 95% CI: 0.54-1.13; fever: RR = 1.19, 95% CI: 0.91-1.56). The PD-1/PD-L1 outperformed comparators in gastrointestinal disorders and blood disorders but failed to do so with the liver system. In gastrointestinal disorders, the incidence of nausea, vomiting, and diarrhea were associated with better improvement in treatment group than that in control, with the risk ratio and 95% CI of 0. [bib_ref] Immune-related adverse events of a PD-L1 inhibitor plus chemotherapy versus a PD-L1..., Wang [/bib_ref]
## Grade 3 or higher adverse event
The incidence of adverse events of grade 3 or higher was low-level both in the treatment group and the control group. [fig_ref] Table 3: Results of Grade 3 or higher adverse events [/fig_ref] summarizes the number of trials, corresponding risk ratio and heterogeneity analysis of grade 3 or higher adverse events using PD-1/PD-L1 inhibitors. It is apparent from this table that PD-1/PD-L1 inhibitors took highly risk in serious dyspnea compared with that of control groups (RR: 1.55, 95% CI: 1.13-2.12). No statistically significant difference was found in the incidence of serious general adverse events (fatigue: RR = 0.78, 95% CI: 0.54-1.13; fever: RR = 1.19, 95% CI: 0.91-1.56). The PD-1/PD-L1 outperformed comparators in gastrointestinal disorders and blood disorders but failed to do so with the liver system. In gastrointestinal disorders, the incidence of nausea, vomiting, and diarrhea were associated with better improvement in treatment group than that in control, with the risk ratio and 95% CI of 0. [bib_ref] Immune-related adverse events of a PD-L1 inhibitor plus chemotherapy versus a PD-L1..., Wang [/bib_ref] 95% CI: 0.06-0.16). However, using PD-1 or PD-L1 inhibitors showed an increased risk of hepatitis with RR equal to 3.45. There were 34 trials used PD-1 inhibitors included camrelizumab, cemiplimab, nivolumab, and pembrolizumab, 11 trials for PD-L1 inhibitors with atezolizumab, avelumab, and durvalumab. Subgroup analysis demonstrated PD-L1 inhibitors associated with higher risk of fever (RR: 1.56, 95% CI: 1.24-1.97; see and headache (RR: 1.55, 95% CI: 1.19-2.91; see For PD-L1 inhibitors versus PD-1 inhibitors, the risk of ALT elevation and AST elevation were both higher (RR: 2.10 versus 1.48; 2.34 versus 1.39, respectively). To better estimate the safety of different checkpoint inhibitors, we then conducted subgroup analysis by single drugs. We found that atezolizumab was also associated with the highest risk of AST elevation (RR: 3.39, 95% CI: 2.26-5.07) and there were 5 trials using atezolizumab reported AST elevation [fig_ref] Figure 5: Subgroup analysis of AST elevation in different immune checkpoint inhibitors [/fig_ref].
## Subgroup analysis by treatment frequency
According to the frequency of treatment, all the trials were divided into two subgroups, 19 trials every 2 weeks (q2w) and 26 trials every 3 weeks (q3w). Except for myalgia, both groups showed consistent results of treatment-related adverse events. The findings indicated that treatment with q2w led to a significant risk in myalgia (RR: 1.48, 95% CI: 1.03-2.15), while a lower risk of myalgia was favored by q3w (RR: 0.71, 95% CI: 0.53-0.97; [fig_ref] Figure 1, Figure 1: Flow diagram of search and study selection [/fig_ref].
## Subgroup analysis by control group
A total of 30 trials used chemotherapy as control group, and 4 trials entered nonchemotherapy. Another 4 trials used ipilimumab and the rest belonged to placebo. When stratifying trials according to control groups, the result of myalgia is quite different between chemotherapy group and other groups. Compared with that of chemotherapy, PD-1/PD-L1 inhibitors were associated with a lower risk of myalgia (RR: 0.56, 95% CI: 0.45-0.70), but compared with other subgroups, the result was in contrary, as is shown in [fig_ref] Figure 6: Forest plot of subgroup analysis of myalgia in different control groups [/fig_ref]. At the same time, immune-related adverse events were of higher risk when compared with that of chemotherapy groups, such as ALT elevation (RR: 1.56, 95% CI: 1.22-1.99), AST elevation (RR: 1.67, 95% CI: 1.33-2.10), and pruritus (RR:2.83, 95% CI: 2.27-3.51). For other treatment-related adverse events, PD-1/PD-L1 inhibitors increased the risk of cough (RR: 1.33, 95% CI: 1.23-1.44) and dyspnea (RR: 1.26, 95% CI: 1.17-1.37) compared with that of chemotherapy, as is shown in Figures S12 and S13.
## Subgroup analysis by treatment frequency
According to the frequency of treatment, all the trials were divided into two subgroups, 19 trials every 2 weeks (q2w) and 26 trials every 3 weeks (q3w). Except for myalgia, both groups showed consistent results of treatment-related adverse events. The findings indicated that treatment with q2w led to a significant risk in myalgia (RR: 1.48, 95% CI: 1.03-2.15), while a lower risk of myalgia was favored by q3w (RR: 0.71, 95% CI: 0.53-0.97; [fig_ref] Figure 1, Figure 1: Flow diagram of search and study selection [/fig_ref].
## Subgroup analysis by control group
A total of 30 trials used chemotherapy as control group, and 4 trials entered nonchemotherapy. Another 4 trials used ipilimumab and the rest belonged to placebo. When stratifying trials according to control groups, the result of myalgia is quite different between chemotherapy group and other groups. Compared with that of chemotherapy, PD-1/PD-L1 inhibitors were associated with a lower risk of myalgia (RR: 0.56, 95% CI: 0.45-0.70), but compared with other subgroups, the result was in contrary, as is shown in [fig_ref] Figure 6: Forest plot of subgroup analysis of myalgia in different control groups [/fig_ref]. At the same time, immune-related adverse events were of higher risk when compared with that of chemotherapy groups, such as ALT elevation
# Discussion
In this study, we performed a systematic review and meta-analysis of treatment-related adverse events of PD-1/PD-L1 inhibitors in randomized clinical trials. We found that in all-grade adverse events, PD-1/PD-L1 inhibitors were associated with significant increase in respiratory disorders like cough, dyspnea, and chest pain. What's more, from our analysis, in addition to focusing on immune-related adverse events, treatment-related adverse events such as arthralgia and myalgia should not be ignored.
# Discussion
In this study, we performed a systematic review and meta-analysis of treatmentrelated adverse events of PD-1/PD-L1 inhibitors in randomized clinical trials. We found that in all-grade adverse events, PD-1/PD-L1 inhibitors were associated with significant increase in respiratory disorders like cough, dyspnea, and chest pain. What's more, from our analysis, in addition to focusing on immune-related adverse events, treatment-related adverse events such as arthralgia and myalgia should not be ignored.
We compared the safety of the targeted PD-1/PD-L1 inhibitors with that of the corresponding control group. The results implied that PD-1/PD-L1 inhibitors were associated with a lower risk of all grade or grade 3 or higher treatment-related adverse events. Fatigue is the most common treatment-related adverse event of PD-1 or PD-L1 inhibitors. In singleagent studies, the incidence of fatigue with anti-PD-1 drugs was 18-26% [bib_ref] Treatment-Related Adverse Events of PD-1 and PD-L1 Inhibitors in Clinical Trials: A..., Wang [/bib_ref]. Normally, fatigue is mild and has nothing to do with other systemic symptoms [bib_ref] Toxicities of the anti-PD-1 and anti-PD-L1 immune checkpoint antibodies, Naidoo [/bib_ref]. Patients treated with PD-1/PD-L1 inhibitors had less likelihood of loss of appetite, nausea, vomiting, and diarrhea, indicating that they may be gastrointestinal-friendly. Meanwhile, immune-related adverse events were frequently reported with PD-1/PD-L1 inhibitors, including rash, pruritus, colitis, aspartate aminotransferase elevation and hypothyroidism [bib_ref] Outcomes Associated with First-Line anti-PD-1/PD-L1 agents vs. Sunitinib in Patients with Sarcomatoid..., Buonerba [/bib_ref] [bib_ref] Safety and Tolerability of PD-1/PD-L1 Inhibitors Compared with Chemotherapy in Patients with..., Nishijima [/bib_ref]. Consistent with De Velasco G's results [bib_ref] Comprehensive Meta-analysis of Key Immune-Related Adverse Events from CTLA-4 and PD-1/PD-L1 Inhibitors..., Velasco [/bib_ref] , our study also found that PD-1/PD-L1 inhibitors were associated with more all-grade rash, AST elevation, and hypothyroidism. Skin rash is the most common adverse reaction associated with immune checkpoint therapy. It may be related to Stevens Johnson syndrome and epidermal necrosis, or it may be the blocking effect of drugs on patients' tumor cells and other common antigens at skin nodes [bib_ref] Treatment-Related Adverse Events of PD-1 and PD-L1 Inhibitors in Clinical Trials: A..., Wang [/bib_ref]. For patients with severe skin diseases, early dermatological diagnosis and evaluation are recommended. What's more, a PD-L1 inhibitor plus chemotherapy may decrease the skin reaction compared with a PD-L1 inhibitor alone [bib_ref] Immune-related adverse events of a PD-L1 inhibitor plus chemotherapy versus a PD-L1..., Wang [/bib_ref]. For further research, we can discuss the adverse events of combined therapy.
The liver damage is mainly manifested in the undifferentiated increase of AST and ALT, and the pathological appearance of induced hepatitis and ipilimumab are similar [bib_ref] Pathologic changes in ipilimumab-related hepatitis in patients with metastatic melanoma, Kleiner [/bib_ref]. In subgroup analysis, we observed higher risk of AST elevation in PD-L1 inhibitors compared with that of nonimmune checkpoint inhibitors. This outcome is contrary to that of Sonpavde et al. [bib_ref] Immune-related adverse events with PD-1 versus PD-L1 inhibitors: A meta-analysis of 8730..., Sonpavde [/bib_ref] who found a lower risk of ALT elevation for anti-PD-L1 inhibitors versus PD-1 inhibitors. Actually, the difference of two meta-analysis may come from the different methods used in meta-analysis. For the previous study, they used indirect incidence rate and were hypothesis-generating [bib_ref] Immune-related adverse events of a PD-L1 inhibitor plus chemotherapy versus a PD-L1..., Wang [/bib_ref] [bib_ref] Immune-related adverse events with PD-1 versus PD-L1 inhibitors: A meta-analysis of 8730..., Sonpavde [/bib_ref] , while our study used randomized controlled trials, so we could calculate the relative risk.
The most important clinically relevant finding was that PD-1/PD-L1 inhibitors were associated with more respiratory events like all grade cough, all grade and grade 3 or higher dyspnea. Previous single-arm trial CheckMate 063 with anti-PD-1 inhibitors reported 5% patients suffering from dyspnea [bib_ref] Activity and safety of nivolumab, an anti-PD-1 immune checkpoint inhibitor, for patients..., Rizvi [/bib_ref]. Another observational study also noted that 1.4% patients developed cough and 0.8% dyspnea after treatment with nivolumab or pembrolizumab [bib_ref] Neurological, respiratory, musculoskeletal, cardiac and ocular side-effects of anti-PD-1 therapy, Zimmer [/bib_ref]. Our results provided evidence that compared with chemotherapy, PD-1/PD-L1 inhibitors increased the risk of cough and dyspnea.
Moreover, the results of this meta-analysis are notable that arthralgia took a higher risk in PD-1/PD-L1 inhibitors (RR = 1.27). Analysis of KEYNOTE-028 trials found that arthralgia was the most common treatment-related adverse events (19.2%) in pembrolizumab [bib_ref] Pembrolizumab in patients with programmed death ligand 1-positive advanced ovarian cancer: Analysis..., Varga [/bib_ref]. Baxi et al. discovered higher rates of musculoskeletal problems in clinical trials treated with PD-1/PD-L1 inhibitors, however, they failed to conduct a meta-analysis because of incomplete data [bib_ref] Immune-related adverse events for anti-PD-1 and anti-PD-L1 drugs: Systematic review and meta-analysis, Baxi [/bib_ref]. To our knowledge, this is the first meta-analysis focused on musculoskeletal disorders. In our study, there were 39 trials all reporting adverse events related to musculoskeletal system, therefore we could do a meta-analysis and then we checked the risk of myalgia and arthralgia. PD-1/PD-L1 inhibitors were associated higher risk of all grade arthralgia (RR = 1.27), however, the pathogenic mechanism was unclear, which call for more molecular biology research. On the other hand, PD-1/PD-L1 decreased the risk of myalgia (RR = 0.56) compared with that of chemotherapy subgroup, but the risk increased with that of other subgroups. Because myalgia is a common adverse event with docetaxel [bib_ref] Taxane-induced arthralgia and myalgia: A literature review, Chiu [/bib_ref] , and thus, in chemotherapy subgroup, the lower risk is acceptable.
Our findings have important implications for clinical oncology treatment choices from the standpoint of patient counseling. From the results of subgroup analysis, we found that the risk of adverse events has a certain difference between PD-1 and PD-L1 inhibitors. If patients choose PD-L1 for cancer treatment, they may face higher risk of fever (RR = 1.56) and headache (RR = 1.55), which may be neglected but equally unbearable compared to that of more serious adverse events. Treatment frequency does not seem to make much difference in the occurrence of adverse events. Our results convinced us that, compared with that of chemotherapy, PD-1/PD-L1 inhibitors reduced the risk of adverse events, especially in digestive disorders and blood system, although they may induce immune-related adverse events. Anyway, when determining a treatment schedule, more consideration can be given to the stage of cancer and patients tolerability, as well as the effectiveness of treatment.
# Limitations
There are some limitations in our meta-analysis. Firstly, the data of adverse events were not taken under uniform standards. The symptoms we analyzed were specialistreported or patient self-reported, such as fatigue, headache, and nausea. In some cases, there were no report of some adverse events and thus missing data were common. Our analysis method could not deal with missing data. Although we tried out best to collect adverse effect data from https://clinicaltrials.gov/ (18 July 2021) where the trials registered, some of our excerpts were still taken from the literature because no results published in website. Inconsistencies in data sources may led to inaccurate analyses. In addition, we compared different controls together, so the results extension was limited. To solve this question, we conducted subgroup analysis. However, in the subgroup analysis, the number of studies in some subgroups was so small that the results were not credible. Finally, considering the above issues, heterogeneity is a serious and non-negligible problem.
# Conclusions
From our meta-analysis, we found that the risk of treatment-related adverse events in PD-1 or PD-L1 varies widely in different systems. In particular, our results provided evidence that PD-1/PD-L1 inhibitors showed higher risk of respiratory disorders including all-grade cough and chest pain, and all-grade and grade 3 or higher dyspnea. What's more, compared with that of chemotherapy, PD-1/PD-L1 inhibitors had lower risk of all-grade myalgia but they were related to higher risk of all-grade arthralgia. These treatmentrelated adverse events should be taken into account when evaluating the safety of immune checkpoint inhibitors.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10.339 0/life11111277/s1, : The risk of bias summary, [fig_ref] Figure 2: Risk of bias graph [/fig_ref] -S13: Forest plots for different adverse events, respectively.
[fig] Figure 1, Figure 1: Flow diagram of search and study selection. PRISMA 2020 flow diagram for new systematic reviews which included searches of databases and registers only PD-L1 used both in treatment and control group (n = 174) Not use PD-1 or PD-L1 alone in treatment (Flow diagram of search and study selection. [/fig]
[fig] Figure 2: Risk of bias graph: two authors' judgements about each risk of bias item presented as percentages across all included studies. [/fig]
[fig] Figure 3: Forest plot of all grade fatigue for PD-1/PD-L1 inhibitors compared with that of control groups. [/fig]
[fig] Figure 4: Forest plot of all grade arthralgia for PD-1/PD-L1 inhibitors compared with that of control groups. [/fig]
[fig] Figure 5: Subgroup analysis of AST elevation in different immune checkpoint inhibitors. [/fig]
[fig] Figure 6: Forest plot of subgroup analysis of myalgia in different control groups. [/fig]
[fig] Funding: This research was funded by the National Natural Science Foundation of China [No.11901013], Beijing Natural Science Foundation [No.1204031], the Fundamental Research Funds for the Central Universities [BMU2021RCZX023], and the Project 2020BD029 supported by PKU-Baidu Fund. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. [/fig]
[table] Table 1: Characteristics of included studies.Table 1. Cont. [/table]
[table] Table 2: Results of all grades' adverse events. [/table]
[table] Table 3: Results of Grade 3 or higher adverse events. [/table]
|
The expression of endogenous voltage‐gated potassium channels in HEK293 cells is affected by culture conditions
HEK293 cells are widely used as a host for expression of heterologous proteins; yet, little care has been taken to characterize their endogenous membrane components, including ion channels. In this work, we aimed to describe the biophysical and pharmacological properties of endogenous, voltage-dependent potassium currents (IKv). We also examined how its expression depends on culture conditions. We used the electrophysiological technique of whole-cell patch clamp to record ion currents from HEK293 cells. We found that HEK cells express endogenous, voltage-dependent potassium currents. We also found that diverse culture conditions, such as the passage number, the cell density, the type of serum that complements the culture media and the substratum, affect the magnitude and shape of IKv, resulting from the relative contribution of fast, slow, and noninactivating component currents. Incubation of cells in mature monolayers with trypsin-EDTA, notoriously reduces the magnitude and modifies the shape of voltage-dependent potassium endogenous currents; nonetheless HEK cells recover IKv 0 s magnitude and shape within 6 h after replating, with a process that requires synthesis of new mRNA and protein subunits, as evidenced by the fact that actinomycin D and cycloheximide, inhibitors of synthesis of mRNA and protein, respectively, impair the recovery of IKv after trypsinization. In addition to be useful as a model expression system, HEK293 may be useful to understand how cells regulate the density of ion channels on the membrane.
# Introduction
HEK-293 is an immortalized cell line, of epithelial morphology, obtained by exposure of human embryonic kidney cells to sheared fragments of human adenovirus type 5 DNA [bib_ref] Characteristics of a human cell line transformed by DNA from human adenovirus..., Graham [/bib_ref]. Given that these cells are very easy to grow and transfect, they have been used as a host model for expression and analysis of diverse heterologous proteins, such as receptors [bib_ref] Morphine activates opioid receptors without causing their rapid internalization, Keith [/bib_ref] [bib_ref] Structure-function relationships of the luteinizing hormone receptor, Puett [/bib_ref] , pumps [bib_ref] Cloning, expression and functional characterization of the putative regeneration and tolerance factor..., Babichev [/bib_ref] [bib_ref] Sarco/endoplasmic reticulum Ca2 + ATPase type 3 isoforms (SERCA3b and SERCA3f): distinct..., Chaâbane [/bib_ref] , and carriers [bib_ref] Comparison of expressed human and mouse sodium/iodide symporters reveals differences in transport..., Dayem [/bib_ref] [bib_ref] A high-capacity membrane potential FRET-based assay for the sodiumcoupled glucose co-transporter SGLT1, Weinglass [/bib_ref] ; They have been used also for production of synthetic polymer nanoparticles [bib_ref] Physicochemical characterization of poly(Llactic acid) and poly(D, L-lactide-co-glycolide) nanoparticles with polyethylenimine as..., Kim [/bib_ref]. Because this permanently transformed cell line has incorporated Ad5 into chromosome 19, it has been used for generation of recombinant E1-deleted human adenoviral vectors [bib_ref] HEK293 cell line: a vehicle for the expression of recombinant proteins, Thomas [/bib_ref] as well as retroviral-based vectors [bib_ref] Scalable production of adenovirus vectors, Silva [/bib_ref].
HEK293 cells are also a preferred choice among electrophysiologists, who seek to study the biophysical properties of heterologous ion channels, due to its small size and voltage can be conveniently clamped. In addition, gigaseals are easily accomplished because HEK cells have no conspicuous processes. Therefore, HEK cells have been used to study the properties of a variety of exogenous ion channels, including sodium channels [bib_ref] Mechanosensitivity of Nav1.5, a voltagesensitive sodium channel, Beyder [/bib_ref] [bib_ref] A missense mutation of the gene encoding voltage-dependent sodium channel (Nav1.1) confers..., Mashimo [/bib_ref] [bib_ref] Biophysical and Pharmacological Characterization of Nav1.9 Voltage Dependent Sodium Channels Stably Expressed..., Lin [/bib_ref] , potassium channels [bib_ref] Kv1.5 open channel block by the antiarrhythmic drug disopyramide: molecular determinants of..., Ar Echiga [/bib_ref] [bib_ref] Inhibition of A-type potassium current by the peptide toxin SNX-482, Kimm [/bib_ref] [bib_ref] Regulation of Kv4.2 A-type potassium channels in HEK-293 cells by hypoxia, Liu [/bib_ref] [bib_ref] Biophysical characterization of inwardly rectifying potassium currents (I(K1) I(K, ACh), I(K, Ca))..., Tang [/bib_ref] , calcium channels [bib_ref] Molecular characterization of two members of the T-type calcium channel family, Perez-Reyes [/bib_ref] [bib_ref] LRRK2 regulates voltage-gated calcium channel function, Bedford [/bib_ref] , chloride channels [bib_ref] Ca(2 + )-activated Cl(-) channels: a newly emerging anion transport family, Fuller [/bib_ref] [bib_ref] Kidney CLC-K chloride channels inhibitors: structure-based studies and efficacy in hypertension and..., Liantonio [/bib_ref] , transient receptor potential channels [bib_ref] Molecular and functional characterization of the melastatin-related cation channel TRPM3. 1, Grimm [/bib_ref] [bib_ref] Cold sensitivity of recombinant TRPA1 channels, Sawada [/bib_ref] [bib_ref] Voltage-and colddependent gating of single TRPM8 ion channels, Fern Andez [/bib_ref] , and aquaporins [bib_ref] Contribution of aquaporins to cellular water transport observed by a microfluidic cell..., Heo [/bib_ref] [bib_ref] Osmosensitivity of transient receptor potential vanilloid 1 is synergistically enhanced by distinct..., Nishihara [/bib_ref].
Despite being so widely used, relatively little attention has been taken to study the variety of endogenous ion channels that these cells express, let alone to understand the factors that determine and regulate its expression. Endogenous channels that had been described so far include voltage-gated sodium and calcium channels [bib_ref] Human embryonic kidney (HEK293) cells express endogenous voltage-gated sodium currents and Na..., He [/bib_ref] , as well as TRP channels [bib_ref] Biochemical, single-channel, whole-cell patch clamp, and pharmacological analyses of endogenous TRPM4 channels..., Amarouch [/bib_ref]. Endogenous voltage-dependent potassium currents have been described in HEK293, although with discrepant results: [bib_ref] Identification of endogenous outward currents in the human embryonic kidney (HEK 293)..., Zhu [/bib_ref] reported that potassium channels have a minor contribution to the observed outward currents. [bib_ref] Endogenous voltage-gated potassium channels in human embryonic kidney (HEK293) cells, Yu [/bib_ref] observed only delayed rectifier but no IA currents. [bib_ref] Intramembrane charge movement associated with endogenous K+ channel activity in HEK-293 cells, Avila [/bib_ref] reported at least two different types of voltage-gated potassium channels, although they did not address their identity. Later on, [bib_ref] Endogenous channels in HEK cells and potential roles in HCN ionic current..., Varghese [/bib_ref] and [bib_ref] Changes in ultrastructure and endogenous ionic channels activity during culture of HEK..., Kurejov A [/bib_ref] reported that both IKand IA-type currents are endogenously expressed in these cells. In this work, we studied the functional and pharmacological properties of endogenous, voltage-dependent potassium currents of HEK293 cells and how culture conditions influence their expression. We also studied how trypsinization of cells during the process of harvesting and subculturing affect these currents.
# Materials and methods
Culture of cells HEK293 cells were cultured in disposable Petri dishes with a medium composed of Dulbecco's modified Eagle's (DMEM, GIBCO, 12800-017), 10% fetal bovine serum (Gibco TM Fetal Bovine Serum, Qualified, Cat. 26140095), 2 mmol/L glutamine, and penicillin-streptomycin 100 U/ mL, and kept in an incubator at 37°C in a humidified atmosphere containing 95% CO 2 . The culture medium was changed every 2 days.
Cells were subcultured once a week, by trypsinization, [Trypsin 0.25% (w/v)-0.53 mm EDTA] followed by gently repipetting, and plated at a density of 4 9 10 3 cells per square centimeter. The culture medium was replaced every 3 days. For electrophysiological recording purposes, cells were plated on glass coverslips, previously placed in 35-mm Petri dishes.
## Electrophysiological recording of cells
Membrane ion currents were recorded using the wholecell patch clamp technique following standard procedures, as described elsewhere [bib_ref] Expression of voltage dependent potassium currents in freshly dissociated rat articular chondrocytes, Ponce [/bib_ref]. Briefly, micropipettes were produced by heating and pulling borosilicate glass tubing (cat. 34500-99, Kimble Chase, USA) with a horizontal puller device (P-87, Sutter Instrument Co. CA, USA). Tip resistance of micropipettes was from 2 to 5 MΩ after heat polishing. Micropipettes were backfilled with a saline solution (see Solutions) and attached through a pipette holder to a piezoelectric-driven micromanipulator (PCS6000, Burleigh Co.). Mechanical manipulation of pipettes was monitored with an inverted microscope (Diaphot 300, Nikon, Japan). Glass coverslips containing HEK-293 cells were immersed on a chamber containing an extracellular solution (see Solutions) and continuously perfused. A glass tubing filled with 2% agarose in 500 mmol/L KCl was set to made electrical contact between the bathing solution and the reference electrode, which was immersed in 500 mmol/L KCl. Voltage pulse protocols and recording of ion currents were made with a patch clamp amplifier (8900, DAGAN Corp, MN, USA) controlled by a dedicated software suite (pclamp 8.0, Axon Instruments Inc. CA, USA). Patch rupture was achieved by suction after gigaseal reached values greater than 2 GΩ (typically 5 GΩ). Unless otherwise stated, membrane potential protocol consisted of series of square pulses, which from a holding value of À120 mV, changed to a test potential from À60 mV to +80 mV in steps of 20 mV, then returning to a holding potential of À80 mV. A p/4 protocol was set to subtract linear components.
## Measurement of membrane capacitance
A capacitive current transient was induced by a hyperpolarizing square pulse of voltage, from À100 to À110 mV, and recorded at 10 KHz. Membrane capacitance was calculated offline by integrating the area of the capacitive transient at the onset of the pulse, then dividing the integrate by the amplitude of the pulse (À10 mV), according to the following equation:
[formula] c m ¼ R 1 t 0 I c ÁDt DV [/formula]
where c m is the membrane capacitance, I c the capacitive current, and DV the amplitude of the voltage pulse (À10 mV). Calculation of the integrate was made with the clampfit module of pClamp 8.0 (Molecular Devices).
## Solutions
Pipette (intracellular) solution was composed of (mmol/ L): 135 K-gluconate, 5 KCl, 1 MgCl 2 , 5 glucose, 10 HEPES, 10 EGTA, pH 7.4, adjusted with KOH. Extracellular solution composition consisted of (mmol/L): 140 Na-gluconate, 5 K-gluconate, 3 CaCl 2 , 1 MgCl 2 , 5 glucose, 10 HEPES, pH 7.4 adjusted with NaOH.
## Chemicals and drugs
All salts, chemicals, and drugs were purchased from Sigma-Aldrich. Actinomycin D (A9415) was dissolved in DMSO (5 mg/mL) prior to use. Cycloheximide (C4859) was obtained as a ready-made solution (100 mg/mL in DMSO).
# Statistical analysis
Descriptive statistics, significance tests, and ANOVA of single factor were made with the analysis module of EXCEL . A minimal level of a = 0.05 was taken as statistically significant.
# Results
## Biophysical properties of endogenous potassium currents of hek293 cells
Ion currents were recorded from HEK cells in mature monolayers by the whole-cell clamp technique. The protocol of stimulation consisted of a series of squared test pulses from À80 to +80 mV, lasting 1000 msec, in steps of 20 mV, from a holding of À120 mV. To exclude anion currents, chloride was substituted by gluconate on the external media composition (see Solutions). Figure 1A shows a representative series of traces of ion currents obtained under these conditions (IKT). Ion currents activate, increasing their magnitude over time until reaching a sustained level. Both the promptness of activation and the sustained magnitude increase with voltage. From about +20 mV, currents start displaying inactivation: Ion currents increase its magnitude over time until reaching a peak and then decline with a kinetics that can be fitted with two time constants, a fast and a slow (as pointed out by arrows on [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref]. This multiple kinetics of the inactivating phase suggests that several functionally distinct components are contributing to shape such currents. This fact prompted us to discern such components with two alternate protocols of stimulation: First, a second series of test pulses was set, similar to the previously described; however, this time each test pulse was led by a prepulse of À30 mV during 500 msec, which was intended to inactivate the fast inactivating component (IKF). This component was further revealed by subtraction of currents, recorded with and without the prepulse [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref]. Second, to discern the slowly inactivating component (IKS), cells were first stimulated with a set of pulses similar to the previous one, but having the time length of each test pulse extended to 10 sec [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref] , bottom); immediately after, cells were stimulated with another series of pulses, of the same amplitude and length, but now each pulse was preceded by a pulse of À30 mV for 1.5 sec. A lapse of 10 sec was set between episodes to allow recovery of slow inactivation. This procedure excluded the inactivating components, leaving only noninactivating currents (IKN) as shown in [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref]. The slowly inactivating currents (IKS, [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref] were further revealed from the difference between traces, as in 1D, from traces as in 1E. [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref] shows the relationship between the average current density (AES.E., pA/pF) and the test voltage from the total potassium current (IKT) as well as from each of its functional components. IKT (cherry circles) has an outward rectification profile that begins to be conspicuous at À60 mV, with a current density of 0.64 AE 0.5 pA/pF up to 16.8 AE 2.0 pA/pF at +80 mV. IKF(orange circles), the fast inactivating component, starts to be noticeable until 0 mV with a magnitude of 4.05 AE 1.1 pA/pF that increases to 11.2 AE 2.3 pA/pF at +80 mV. IKS (light green triangles), the slowly inactivating component, starts to be manifest at À20 mV with 1.85 AE 0.8 pA/pF rising to 6.5 AE 1.2 pA/pF at +80 mV. Finally, IKN (dark green triangles), the noninactivating component, starts to be noticeable from around À40 mV with 1.7 AE 0.5 pA/pF up to 7.3 AE 1.1 pA/pF at +80 mV. [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref] shows the relationship between the mean conductance density (pS/pF) and the test potential of all distinct components of current. These data were fitted with the Boltzmann equation:
[formula] g ¼ g 0 þ g max 1þe À vÀv 1=2 k À Á [/formula]
where g max is the maximal conductance, v 1/2 is the test potential producing half conductance activation, and k is a parameter determining the stepness of voltage dependence. IKT was fitted with g max of 75 AE 2 pS/pF and a v 1/2 of À10.3 AE 1.3 mV; IKF was fitted with g max of 49 AE 2.9 pS/ pF and a v 1/2 of À3.1 AE 2.2 mV, b 10.8; For IKS, g max was 28 AE 1.9 pS/pF and v 1/2 = À21.8 AE 3.3 mV. IKN was fitted with g max of 37 AE 6.6 pS/pF and v 1/2 of À45.2 AE 9.5 mV.
For each current component, the time constant of activation (s on ) was obtained by fitting the rising phase of current traces at each test voltage. [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref] shows that for all the three components of current (IKF, IKS, and IKN), the relationship between s on and the test voltage follows an exponentially decaying trend. For IKF, s on ranges from 253 AE 50 msec at À80 mV to 4 AE 2 msec at +80 mV; for IKS, 186 AE 50 msec to 26 AE 5 msec; and for IKN 10032 AE 120 msec to 68 AE 50 msec. The time constant of inactivation (s inact. ) was obtained by fitting the decaying phase of current traces [fig_ref] Figure 1: Endogenous ion currents from HEK293 cells [/fig_ref]. It ranged from 1480 AE 220 to 46 AE 10 msec for IKF, whereas for IKS it was from 7212 AE 220 to 7000 AE 90 msec, which resulted to be nonsignificantly distinct (P > 0.05, ANOVA one-way).
In order to verify the selectivity of these currents, we designed a protocol, aimed to induce tail currents: starting from a holding of À80 mV, voltage was switched to +40 mV for 1 sec, to activate IKT currents, then to a varying test voltage from À120 to À40 mV in steps of 10 mV. From the relationship between the initial magnitude of tail currents and the test voltage, the reversal potential (Er) was interpolated. As [fig_ref] Figure 2: Potassium selectivity of endogenous ion currents from HEK293 [/fig_ref] (A and B) shows, Er shifts from À70 mV when external potassium concentration is 3 mmol/L to À48 mV when potassium is raised to 9 mmol/L. These results, therefore, indicate that such currents are selective to potassium. [fig_ref] Figure 2: Potassium selectivity of endogenous ion currents from HEK293 [/fig_ref] shows the relationship between the time constant of deactivation and the test voltage. It follows a simple, exponential trend, and ranges from 1.9 AE 0.5 msec at À120 mV to 11 AE 2 msec at À40 mV.
## Pharmacological properties of endogenous k currents of hek-293 cells
We analyzed the effect of TEA and 4AP, which are general K channel blockers, on IKT as well as on each one of its components. For this purpose, we recorded current traces, induced by test pulses of +60 mV, from cells incubated with progressively increasing concentrations of blockers in the external media. As [fig_ref] Figure 3: Pharmacological properties of endogenous K currents of HEK-293 cells [/fig_ref] shows, both compounds effectively reduced the size of total K currents as well as its components, although with distinct sensitivity. To quantify such sensitivity, a percent blocking index (%B d ) was calculated for each drug concentration with the following transformation:
[formula] %B d ¼ I0ÀI d I 0 Â 100 [/formula]
Where I d is the peak magnitude of ion currents at a given drug concentration, whereas I 0 is the corresponding value with no blocker added. Plotting of these values versus the log 10 of blocker concentration follows a sigmoidal relationship that was fitted with the following function:
[formula] %B x ¼ Bmax 1þe À xÀx 50 b ð Þ [/formula]
where B max is the maximal blocking effect of the drug, x is a given testing drug concentration, and x 50 the concentration blocking half the amplitude of currents. TEA produced a B max of 91% on IKT with an x 50 of 2.9 AE 0.5 mmol/L (n = 10); On IKF, B max was 95% with x 50 of 1.0 AE 0.005 mmol/L (n = 9); a similar result was found for IKS, with B max of 94% and x 50 of 1.5 AE 0.1 mmol/L (n = 9). IKN was blocked with B max of 94% and x 50 of 4.19 AE 0.05 mmol/L (n = 10). 4AP also blocked IKT currents, although it produced a lower maximal effect (68%) than TEA, a lower concentration was needed to produce a half effect(x 50 = 0.3 AE 0.08 mmol/L, n = 10); It blocked more efficiently IKF (B max = 97.5%, x 50 = 0.08 AE 0.005 mmol/L, n = 9) than IKS (B max = 70%, x 50 = 0.37 AE 0.01 mmol/L, n = 9) and IKN (B max = 55.8%, x 50 = 0.37 AE 0.01 mmol/L, n = 10).
In addition to TEA and 4-AP, we examined the effect of a set of toxins that has been described as specific blockers of molecular entities of voltage-dependent potassium channels of the Kv1 subfamily: a-dendrotoxin targets Kv1.1, Kv1.2, and Kv1.6 (Harvey 2001); noxiustoxin, a potent blocker of Kv1.2 and Kv1.3; charybdotoxin, a potent blocker of K Ca 1.1, Kv1.2, and Kv1.3 ; agitoxin-1, which targets Kv1.3 [bib_ref] Purification and characterization of three inhibitors of voltage-dependent K+ channels from Leiurus..., Garcia [/bib_ref] ; and margatoxin, a specific blocker of KV1.3 and KV1.6 (Leonard et al. 1992; [bib_ref] Purification, characterization, and biosynthesis of margatoxin, a component of Centruroides margaritatus venom..., Garcia-Calvo [/bib_ref]. We added those toxins (a single concentration) to the external solution and compared the magnitude of the peak current at +60 mV before and after its addition. [fig_ref] Figure 4: Pharmacological properties of endogenous K currents of HEK-293 cells [/fig_ref] shows a representative example of the effect of these toxins on IKT as well as on its functional components. [fig_ref] Figure 4: Pharmacological properties of endogenous K currents of HEK-293 cells [/fig_ref] shows the averaged % blocking effect that these toxins produce on each functional component. 2018 | Vol. 6 | Iss. 8 | e13663
Page 5
Influence of culture conditions on endogenous K currents Next, we evaluated whether culture conditions affect the expression of endogenous K currents. For this purpose, we analyzed how these currents are modified by changes in the passage number, the cell density, the substrate, and the serum complementing the culture media. In order to compare the magnitude of currents at distinct values of each condition, we recorded IKT as well as IKF, IKS, and IKN currents in response to a test pulse of +60 mV from a number of HEK293 cells to make statistical analysis.
## Passage number
We compared K currents from cells at passage numbers 20, 30, 50, and 70. As shown on [fig_ref] Figure 5: Influence of culture factors on endogenous K currents of HEK-293 cells [/fig_ref] , we found that IKT increase significantly (P < 0.001, ANOVA) from 11.8 AE 1.7 pA/pF at passage 20 to 20.8 AE 1.4 pA/pF at passage 70. This increase seems to be mainly due to IKF, whose magnitude increased significantly (P < 0.005, ANOVA) from 6.26 AE 1.23 pA/pF at passage 20 to 13.5 AE 1.1 pA/pF at passage 70. The slow inactivating and the noninactivating currents did not change significantly.
## Serum containing media
We compared K currents from cells grown on culture media complemented with fetal bovine serum (Gibco TM Fetal Bovine Serum, Qualified, Cat. 26140095) versus that from cells grown on media complemented with calf serum (HyClone TM Newborn Calf Serum, GE Healthcare Life Sciences, cat. # SH30118.03). As [fig_ref] Figure 5: Influence of culture factors on endogenous K currents of HEK-293 cells [/fig_ref] shows, the IKT average current density of cells cultured with calf serum (9.6 AE 0.7 pA/pF) was significantly smaller (P < 0.05) than that of cells cultured with fetal bovine serum (12.4 AE 0.9 pA/pF). IKF was also significantly lower (P < 0.01) in cells grown with calf serum (2.7 AE 1.1 pA/ pF) than in cells grown with fetal serum 7.0 AE 1.2 pA/pF, whereas IKS and IKN were not significantly different.
## Cell density
We compared cells seeded at low (5 9 10 2 cells/cm 2 ) versus high density (3 9 10 4 cells/cm 2 ). Both batches of cells were harvested from the same flask, counted, and plated separately. Ion currents were recorded 24 h after seeding [fig_ref] Figure 5: Influence of culture factors on endogenous K currents of HEK-293 cells [/fig_ref]. We found that the membrane surface of cells seeded at high density was significantly (P < 0.01) lower (14.0 AE 0.3 pF) than that of cells seeded at low density (17 AE 0.5 pF). Likewise, the average IKT current density was significantly smaller from cells plated at low density than from cells plated at high density (P < 0.05). The mean current density of all three functional components (IKF, IKS, and IKN) was significantly smaller from cells seeded at high density than from cells seeded at low density.
## Substrate
Endogenous K currents, as well as membrane surface, of cells plated on plastic were compared with that of cells plated on glass. For this purpose, cells were seeded on 60mm Petri dishes containing glass coverslips at 4 9 10 3 cells/cm 2 . Recordings were made 24 h after plating, either on cells deposited on glass coverslips or on the surface of the Petri dish. As [fig_ref] Figure 5: Influence of culture factors on endogenous K currents of HEK-293 cells [/fig_ref] shows, the membrane surface of cells seeded in plastic was significantly (P < 0.05) larger than in cells seeded in glass. Likewise, the IKT current density of cells seeded on a plastic was significantly higher (P < 0.05) than that of cells plated on glass. A significant increase (P < 0.05) was observed in the noninactivating component, whereas no significant difference was observed, neither in IKF nor in IKS.
## Influence of cell-cell contact on the expression of hek endogenous voltagegated potassium channels
To determine if cell-cell contact influences the expression of voltage-dependent potassium channels, we made whole-cell recordings of cells that were either isolated or in contact with neighboring cells and compared the magnitude of IKT currents, as well as its components (IKF, IKS, and IKN). Recordings were made 6 h after plating. Touching and no touching cells were recorded alternatively from a same coverslip containing seeded cells. [fig_ref] Figure 6: Influence of cell-cell contact on endogenous K currents of HEK-293 cells [/fig_ref] shows representative series of currents of each condition. As [fig_ref] Figure 6: Influence of cell-cell contact on endogenous K currents of HEK-293 cells [/fig_ref] shows, we found no significant difference, neither in the mean value of the membrance surface nor in IKT or any of its components. These results suggest therefore that cell-cell contact does not influence the expression of HEK293 endogenous voltagegated potassium currents.
## Effect of trypsinization on endogenous k currents
In order to subculture, HEK293 cells, as most cultured cells, are typically incubated with trypsin-EDTA followed by gentle mechanical dissociation. Because of this procedure, cells detach from its substrate, acquiring a spherical shape, but after seeding, they reattach to the substrate and resume its shape. To evaluate how this treatment affects K currents, we recorded ion currents from cells, as early as 10 min after trypsinization and at subsequent times up to 24 h. [fig_ref] Figure 7: Effect of trypsin on endogenous K currents of HEK-293 cells [/fig_ref] shows representative series of each current component, whereas [fig_ref] Figure 7: Effect of trypsin on endogenous K currents of HEK-293 cells [/fig_ref] shows the time course of the membrane capacity and the density of current of each component (at +60 mV) before (at time 0) and up to 24 h after trypsinization. After treatment, the membrane capacitance of cells recently plated is significantly reduced (P < 0.05). IKT currents also change its magnitude and shape, as they lose the transient peak that is typically recorded from cells before tripsinization. Nonetheless, both features are recovered progressively over time and, after about 24 h of trypsinization, IKT currents recover its original shape and size. 2018 | Vol. 6 | Iss. 8 | e13663
Page 7
In addition to IKT, we analyzed how trypsinization affects each one of the functional components of K currents. IKF decreases significantly (P < 0.001) from 6.8 AE 2.0 (n = 14) before to 0.5 AE 0.1 (n = 12) pA/PF at 10 min after trypsinization; a significant reduction is also observed on IKS (P < 0.05), albeit not as pronounced as IKF, from 5.1 AE 1.2 (n = 13) to 2.3 AE 0.8 (n = 10) pa/PF. IKN did not change significantly from 4.4 AE 1.2 (n = 14) to 3.0 AE 0.8 (n = 12) pA/pF. It is clear from these observations that the subculturing procedure reduces the size and changes the profile of endogenous K currents, affecting most notoriously IKF, the fast inactivating component and to a less degree IKS, the slow inactivating one.
This reduction could be simply due to retrieval of the plasma membrane during the harvesting process but, if that were the case, the density of the channels would be expected to remain unchanged. To probe this possibility, we discarded trypsin out of the harvesting process, detaching cells only by gentle repippeting, and assayed how the membrane surface and endogenous K currents are affected. As [fig_ref] Figure 8: Mechanical detachment of cells does not reduce current density [/fig_ref] shows, although membrane surface was significantly reduced by mechanical harvesting, no significant reduction was observed, neither in the average density of IKT nor in any of its components. Therefore, our results lead us to think that trypsin is selectively degrading some types of potassium channels from those that produce endogenous currents.
## Effect of actinomycin d and cycloheximide on the recovery of surface membrane and potassium currents after tripsinization
To find out whether new RNA and/or protein is required for recovering of K currents after trypsinization, we studied the effect of actinomycin D (5 lg/mL) and cycloheximide (10 lg/mL), drugs that inhibit the synthesis of RNA [bib_ref] Characteristics of a human cell line transformed by DNA from human adenovirus..., Graham [/bib_ref] [bib_ref] Kidney CLC-K chloride channels inhibitors: structure-based studies and efficacy in hypertension and..., Liantonio [/bib_ref] and protein [bib_ref] Actinomycin and DNA transcription, Sobell [/bib_ref] [bib_ref] Inhibition of eukaryotic translation elongation by cycloheximide and lactimidomycin, Schneider-Poetsch [/bib_ref] , respectively. For this purpose, cells in mature monolayers were trypsinized and inhibitors were added to the culture medium immediately after plating. The membrane capacitance and density of current of total (IKT) and each component (IKF, IKS, and IKN), from both control and treated cells, were measured, 5 hours after trypsinization. Cells treated with actinomycin D recovered partially its shape and adherence [fig_ref] Figure 9: Recovery of K currents after trypsinization requires synthesis of new channels [/fig_ref] , yet its membrane surface was significantly lower (P < 0.01) than control cells [fig_ref] Figure 9: Recovery of K currents after trypsinization requires synthesis of new channels [/fig_ref]. This treatment impaired significantly (P < 0.01) the recovery of total endogenous K currents, as well as each of its components: IKF, IKS, and IKN. [fig_ref] Figure 8: Mechanical detachment of cells does not reduce current density [/fig_ref] , upper right). Cycloheximide produced a more dramatic effect than actinomycin D: most cells remained spherical and barely attached to the substratum [fig_ref] Figure 8: Mechanical detachment of cells does not reduce current density [/fig_ref] , this was reflected in a significantly lower membrane surface, as compared to control cells [fig_ref] Figure 8: Mechanical detachment of cells does not reduce current density [/fig_ref]. The recovery of the density of IKT currents was more severely impaired (P < 0.001) by cycloheximide than by actinomycin D [fig_ref] Figure 8: Mechanical detachment of cells does not reduce current density [/fig_ref] , bottom right), mostly due to the fast (IKF) component, whose recovery was significantly impaired (P < 0.01), whereas the recovery of the other components was not significantly affected.
# Discussion
HEK293 is a cell line of striking interest because is used as a model to investigate the functional properties of exogenous proteins, including ion channels. In this work, we describe the biophysical and pharmacological properties of HEK 0 s endogenous, voltage-dependent potassium currents, and show that its magnitude and shape depend on culture conditions, and that the treatment of cells with trypsin, notoriously reduces those currents. It is worth noticing that HEK cells express A-type K currents, which are commonly expressed in excitable cells, such as neurons and cardiomyocytes. Although HEK293 are regarded as epithelial, given its origin from human embryonic kidney [bib_ref] Characteristics of a human cell line transformed by DNA from human adenovirus..., Graham [/bib_ref] , new evidence has emerged suggesting that these cells have instead a neuroendocrine origin derived from the adrenal gland [bib_ref] Preferential transformation of human neuronal cells by human adenoviruses and the origin..., Shaw [/bib_ref] [bib_ref] Characterization of endogenous calcium responses in neuronal cell lines, Vetter [/bib_ref]. Our finding that HEK cells express A-type K currents reinforces such suggestion. We show in this work that the magnitude and shape of endogenous potassium currents are related to culture factors such as the passage number, the media composition, and the substrate on which cells are seeded. This in part may account for the discrepancy in the shape of endogenous currents that has been reported previously [bib_ref] Endogenous voltage-gated potassium channels in human embryonic kidney (HEK293) cells, Yu [/bib_ref] [bib_ref] Identification of endogenous outward currents in the human embryonic kidney (HEK 293)..., Zhu [/bib_ref] [bib_ref] Intramembrane charge movement associated with endogenous K+ channel activity in HEK-293 cells, Avila [/bib_ref] [bib_ref] Endogenous channels in HEK cells and potential roles in HCN ionic current..., Varghese [/bib_ref] [bib_ref] Changes in ultrastructure and endogenous ionic channels activity during culture of HEK..., Kurejov A [/bib_ref]. We found that A-type currents was the most sensitive component, depending on the passage number, as well as the media containing serum and cell density at plating, whereas the delayed rectifier-type currents were less sensitive to such [bib_ref] Endogenous Kv channels in human embryonic kidney (HEK-293) cells, Jiang [/bib_ref]. We also showed that after incubation with trypsin-EDTA, HEK cells reduce its membrane surface as well as the density of K currents, more notoriously the fast inactivating component. The fact that a significant reduction was not observed when cells were detached by simple mechanical enforcement indicates that trypsin degrades an important amount of membrane K channels. We found also that cells restitute both plasma membrane and K currents within a few hours and, the fact that such restitution is abolished by actinomycin D and cycloheximide, suggest that they have to synthesize new channels to restore its former conditions. This process is interesting, because it provides a way to reduce the amount of endogenous K channels, hence minimizing the chance of unexpected results when HEK293 are used for heterologous expression assays. In addition, the fact that HEK cells restitute such channels after trypsinization indicates that HEK are able to regulate the density and variety of its endogenous channels; and the current density of endogenous K currents and its components, and how they recover over time. Red and yellow dots indicate a statistically significant difference of (P < 0.005) and (P < 0.05), respectively. therefore, they could be a model to study the cellular and molecular mechanisms involved in the regulation of the expression of membrane channels. In summary, HEK293 cells have been taken as a host for testing the expression of heterologous proteins, whereas its endogenous components had been generally neglected, however, as is clear from this study, endogenous voltagedependent K currents are not to be neglected, as they could interfere with heterologous K channels. 2018 | Vol. 6 | Iss. 8 | e13663
Page 11
[fig] Figure 1: Endogenous ion currents from HEK293 cells. (A) Representative series of currents recorded from HEK293 cells in response to a voltage protocol as shown in the lower part. (B) Series of currents obtained with a protocol that includes a À30 mV prepulse. (C) Fast inactivating currents (IKF) are obtained by subtraction of traces as in B from traces as in A. (D) Currents recorded with the same protocol as in B but with the length of the pulse extended from 1 to 10 sec show slow inactivation. (E) A prepulse of À30 mV, lasting 1500 msec, reveals the noninactivating component. (F) The slowly inactivating currents are disclosed by subtracting current recordings as in D from those as in E. (G) Relationship between the current density and the test voltage is shown for IKT as well as for each of the functionally distinct components (IKF, IKS and IKN). (H) Relationship between the conductance and the test voltage for IKT and its components. (I) Voltage dependence of the activation kinetics for the distinct K current components. (J) Voltage dependence of the inactivation process for the fast and the slow inactivating components. The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of The Physiological Society and the American Physiological Society. [/fig]
[fig] Figure 2: Potassium selectivity of endogenous ion currents from HEK293. (A) representative examples of tail current series, induced by the voltage protocol shown on the inset, to estimate the reversal potential under distinct external potassium concentrations. (B) The I-V relationship is shifted by changes in the external potassium concentration, as expected for a K selective conductance. (C) Voltage dependence of the time constant of deactivation. ª 2018 The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of The Physiological Society and the American Physiological Society. [/fig]
[fig] Figure 3: Pharmacological properties of endogenous K currents of HEK-293 cells. Effect of TEA and 4AP. (A and B) Representative recordings showing the effect of TEA (left column) and 4AP (right column) on endogenous potassium currents IKT, and its components (IKF, IKS, and IKN). All currents were obtained by test pulses of +60 mV. (C and D) Semilog plots showing the dose-effect relationship. The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of The Physiological Society and the American Physiological Society. [/fig]
[fig] Figure 4: Pharmacological properties of endogenous K currents of HEK-293 cells. Effect of Kv1 blockers. (A) Representative recordings of currents at +60 mV (IKT,IKF,IKS, and IKN) before and after addition of toxins. (B) Statistical analysis showing the percentage blocking effect of each toxin. ª 2018 The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of The Physiological Society and the American Physiological Society. [/fig]
[fig] Figure 5: Influence of culture factors on endogenous K currents of HEK-293 cells. (A) Statistical analysis, comparison of the magnitude of endogenous K currents at +60 mV from cells at passage numbers 20, 30, 50, and 70. The line above groups of bars IKT and IKF denotes a significant difference among bars of the same group (P < 0.001, ANOVA). (B) Comparison of the magnitude of endogenous K currents at +60 mV of cells incubated with fetal serum and cells incubated with calf serum. (C) Comparison of the magnitude of endogenous K currents at +60 mV of cells plated at high versus low density. (D) Comparison of the magnitude of endogenous K currents at +60 mV of cells plated on glass coverslips and of cells plated on the surface of the Petri dish. A (*) above bars denotes a statistically significant difference of (P < 0.05). The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of The Physiological Society and the American Physiological Society. [/fig]
[fig] Figure 6: Influence of cell-cell contact on endogenous K currents of HEK-293 cells. (A) Representative series of IKT, IKF, IKS, and IKN currents from touching and nontouching HEK cells. (B) Comparison of the mean values of each potassium current component (IKT, IKF, IKS, and IKN) and membrane capacitance shows no statistically significant difference. [/fig]
[fig] Figure 7: Effect of trypsin on endogenous K currents of HEK-293 cells. (A) Set of five columns showing (from top to bottom) a representative image of cells in culture at distinct times, before and after treatment with trypsin. (B) Representative series of currents (IKT, IKF, IKS, and IKN) at distinct times after trypsin treatment. (C) Plots showing how trypsin treatment decrease the average value of membrane capacitance (top) [/fig]
[fig] Figure 8: Mechanical detachment of cells does not reduce current density. Graph shows the average value of membrane capacitance (Cm) and endogenous currents before and 15 min after mechanical detachment of cells, without trypsin. As shown, this treatment did not change the current density but the membrane capacitance. The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf ofThe Physiological Society and the American Physiological Society. [/fig]
[fig] Figure 9: Recovery of K currents after trypsinization requires synthesis of new channels. (A) Representative images of cultured HEK cells, obtained 5 h after trypsinization and seeding. (B) Bars comparing the membrane capacitance (left) and current density of IKT, IKF, IKS, and IKN under control conditions versus treatment with actinomycin D (upper) and cycloheximide (lower). *, **, and *** denotes a statistically significant difference (P < 0.05, 0.01, and 0.005), respectively. ª 2018 The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of The Physiological Society and the American Physiological Society. [/fig]
|
Assessing service and treatment needs and barriers of youth who use illicit and non-medical prescription drugs in Northern Ontario, Canada
BackgroundIllicit drug use rates are high among Canadian youth, and are particularly pronounced in Northern Ontario. The availability and accessibility of effective substance use-related treatments and services are required to address this problem, especially among rural and remote Northern communities. In order to assess specific service and treatment needs, as well as barriers and deterrents to accessing and utilizing services and treatments for youth who use illicit drugs in Northern Ontario, we conducted the present study.
# Methods
This study utilized a mixed-methods design and incorporated a community-based participatory research approach. Questionnaires were administered in conjunction with audiorecorded semi-structured interviews and/or focus groups with youth (agedwho live in Northern Ontario and use illicit drugs. Interviews with 'key informants' who work with the youth in each community were also conducted. Between August and December 2017, the research team traveled to Northern Ontario communities and carried out data collection procedures.
# Introduction
Illicit drug use is a major public health concern worldwide, with an estimated 5.6% of the global population reporting use in 2016, and approximately 31 million of these users meeting requirements for a drug use disorder diagnosis. Illicit drug use rates are commonly elevated among youth. For example, in Canada, 20% of general population adolescents aged, and 35% of those aged 20-24, reported past-year use of illicit drugs in 2017. Moreover, the majority of people who use illicit drugs in Canada are between the ages of 15 to 24, and those who fall into this age category have been found to use drugs more heavily and in riskier ways compared to their older counterparts. Illicit drug use rates among youth in Canada vary across the provinces, with Ontario consistently showing high use rates among adolescents compared to many other provinces. In particular, non-medical prescription opioid use among adolescents has been associated with a variety of adverse health outcomes, including fatal overdoses; there were 240 opioid-related overdose deaths among those aged 24 or younger in Ontario between 2013-2016, 92% of whom did not have an active opioid prescription.
Not only do drug use rates vary across the provinces, they also vary within. For instance, significant regional differences in substance use rates among youth have been found in Ontario (e.g., between the province's Local Integrated Health Networks (LHINs)) (see S1 Appendix), with higher rates of certain substances (e.g., alcohol) reported in Northern Ontario. Prescription opioid use in particular has been found significantly higher among youth in Northern Ontario; accordingly, in 2014, the rate of methadone maintenance treatment patients per 100,000 among youth (ageswas approximately 2-fold and 6-fold higher among the North East and North West LHIN, respectively, compared to the other LHINs in Ontario.
In order to address illicit and non-medical prescription drug use among youth in Northern Ontario, it is imperative that appropriate and culturally relevant treatment options and services are available. There are a number of community-based addiction services in Northern Ontario, which include public, private, and not-for-profit organizations and services, as well as indigenous/culturally-based treatment centers. In terms of opioid agonist treatment for opioid use disorders, evidence suggests that the availability of these programs has recently increased in Ontario, particularly in rural and Northern areas where it had typically been underutilized.
Distinct challenges in providing adequate treatment and service provision in rural and remote communities have been identified. For instance, utilization of addiction services has been found lower in rural compared to urban areas, while access to care has been described as a major challenge since resources are concentrated in cities or larger communities and there are often shortages of healthcare providers. In 2017, youth under the age of 25 living in the North West LHIN had the highest rates of emergency department visits for addiction issues compared to the rest of Ontario, which suggests that there is a lack of access to, or awareness of community-based treatment options.
With this in mind, it is unclear whether or not the addiction-related services that exist in Northern Ontario are being accessed and utilized by the youth who need them, or whether or not the services are effectively addressing their needs. As such, gaining a thorough understanding of the service and treatment needs of youth who use drugs in Northern Ontario communities is essential for identifying and addressing lapses in current service availability and utilization. However, this information is lacking.
Thus, in order to explore the gaps and opportunities in service and treatment needs of youth who use illicit drugs in Northern Ontario, our research team conducted a study examining these issues utilizing a tailored mobile research lab. The research lab enabled the team to conduct data collection processes across multiple locations in a standardized manner, which was essential since Northern Ontario communities are extremely remote, hard to access and considerably dispersed from one another.
The present contribution sheds light on drug use patterns and drug-related health outcomes of youth who use illicit and non-medical prescription drugs in Northern Ontario. We examined needs, major barriers and deterrents, as well as gaps in terms of services and treatments for drug use among this unique population. This information can be used to better meet the needs of youth as well as provide suggestions for service system improvement in Northern Ontario as well as remote and/or rural communities more generally.
# Methods
The registered study protocol and further details on the methodology can be found in Varatharajan et al. (2018). The study was approved by the Centre for Addiction and Mental Health Research Ethics Board (REB# 133/2015).
## Aims and objectives
The objectives of this study were to identify any major gaps in service provision based on the articulated needs of the youth and key informants who reside in Northern Ontario communities, as well as assess barriers and deterrents to service and treatment access and utilization.
## Setting
## Design
The study utilized a convergent-parallel mixed-methods design, employing standardized questionnaires and semi-structured one-on-one interviews and focus groups. We followed a community-based participatory research approachwhere we built and fostered relationships with community members and key informants from service organizations prior to, and maintained these relationships throughout, the data collection phase. Specifically, for each community, we established a main contact/partner/stakeholder (a 'key informant') who was involved throughout the entirety of the project. This contact referred other community members and stakeholders from a wide variety of backgrounds and organizations including people with lived/living experience as well as Indigenous scholars and partners. We collaborated with these partners and received ongoing feedback and input on our research design, operational plans, recruitment, etc., as well as elicited recommendations and suggestions for data collection and analysis in each community.
## Eligibility criteria
The study had two separate types of participants who could be eligible (see protocol for detailed criteria: 1) youth, aged 14-25, who have used illicit and/or non-medical prescription drugs for at least 3 months, on at least 10 days in the past month; and 2) key informants who held a formal position in the community where they worked within a service or organization that served, interacted or worked directly with youth who use illicit and/or nonmedical drugs. Key informants were comprised of people from a variety of backgrounds and organizations, e.g., nurses and other frontline healthcare staff, outreach workers, peer workers, shelter managers, probation and police officers, Indigenous organization directors, mental health and addiction caseworkers, drop-in center managers, community AA/NA sponsors, members from community drug strategy committees or advisory boards, people with lived and living experience, etc.
## Recruitment
Prior to entering the field, the research team utilized our main key informant within each community to help identify additional health and social service organizations/stakeholders who were willing and able to assist with recruitment once we were in their community. We informed these key informants about the aims and objectives of the study via a standardized email letter. Two researchers visited all organizations (including additional organizations identified in the field) in person to meet key informants and to share our recruitment materials with any newly identified organizations. Following a snow-ball sampling approach, key informants were asked to inform their colleagues, community partners and the youth they work with about the study and provide them with posters and handouts containing study details. Key informants were also asked to assist with organizing focus groups who would meet the research team and undergo data collection procedures at pre-established times. The key informants also assisted the research team with identifying a parking spot in each community for the mobile lab that would be visible and easily accessible, and were asked to participate in a one-on-one interview as a key informant participant.
Beyond utilizing key informants to assist with recruitment, while in each community the study team also directly recruited participants by verbally informing individuals who approached the mobile research lab about the study, and by using a respondent-driven sampling technique where we offered participants an additional incentive to refer eligible peers (see protocol for details.
## Study procedures
Between August and December 2017, two researchers from the study team travelled to each community using the mobile research lab. The lab was parked for approximately one week in each community, depending on participant recruitment success.
Potentially eligible participants were either sent to the mobile lab by key informants who had explained the study to them, were identified by approaching the mobile lab and expressing interest in participating, or were recruited by their peers who had completed data collection procedures.
All potentially eligible participants were screened for eligibility, and underwent a written informed consent process (see protocol for eligibility screener and consent form. Parental consent was not sought for minors (agedwho participated in the study as they may have been street-involved, disconnected from their parents, or did not want to reveal their drug use to their parents; this approach was line with legal requirements in Ontario, as long as the participants could be considered a mature person, which was approved by CAMH's Research Ethics Board. After assessing maturity, eligible participants were asked to participate in an anonymous semi-structured, audio-recorded focus group (with three to five participants) or interview (when only one or two participants were eligible at a time). After the focus group/ interview session, participants partook in a self-administered questionnaire (see protocol for quantitative and qualitative assessment tools. At the end of the study session, participants were compensated with a $20 gift card and were provided with a list of available addiction services and treatment programs in their community.
The researchers also conducted one-on-one interviews with key informants, either in person or over the phone; Key informants were not provided any compensation.
## Data collection and analysis
In order to maintain confidentiality and anonymity, all participants were provided with a unique participation study code that was utilized to identify them, and all personal identifiers were removed from the data.
Both quantitative and qualitative data were merged, analyzed and interpreted to inform the results. In order to ensure the results were adequately represented and interpreted, an iterative feedback process was undertaken after initial analyses of our results, and key informants were given the opportunity to review the final research findings prior to publication.
Quantitative analysis. For the quantitative data obtained from the questionnaire responses, basic descriptive analyses were conducted to characterize the study sample. Using simple frequency and cross-tabulations we examined sociodemographic information, drug use patterns and profiles, and health status/diagnoses. All analyses were conducted using Graph-Pad Prism (version 8).
Qualitative analysis. For the qualitative data, audio recordings were transcribed verbatim and imported into qualitative data management software (NVivo, version 12). All interview transcripts underwent an inductive thematic analysis whereby key themes were identified and subsequently coded into categories. Initial themes were developed based on our research questions, and discussed and agreed upon as a group. The team developed an original code list based on these themes, and coding was then conducted by two research analysts (CR and MN); additional codes were added based on emergent themes during analyses. When analyzing/comparing the data between the youth and key informants, data saturation was discussed among the two interviewers, and we conducted a crosswalk to ensure a consensus was reached. Codes and themes were included in final analyses when they were introduced or discussed by numerous participants, and from these, select illustrative samples (from both youth and key informants) were chosen to narratively represent the themes.
# Results
Across the 11 Northern Ontario communities visited, there were a total of n = 102 youth, and n = 35 key informants, who completed the qualitative component of our study; n = 100 youth completed the quantitative component.
# Quantitative results
Sociodemographic information. Participants ranged in age from 14 to 25, with a mean age of 22.2, and were split relatively equally between males (49%) and females (47%). The vast majority (75%) self-identified their ethnic background as 'Aboriginal'. Most participants (39%) reported living at their own place. Regarding highest level of education, half (50%) reported they did not (or had yet to) finish high school. As for sources of income in the last month, 61% of participants reported receiving government assistance (welfare and/or disability) as their primary source of income.
Drug use patterns and profiles. Besides alcohol, tobacco and cannabis, the most commonly used drugs (reported as 'use on more than 10 days in the last month') were any prescription opioids (66%), cocaine (65%), crack-cocaine (52%), benzodiazepines (28%), methamphetamines/crystal meth (24%) and amphetamines/speed/ADHD medication (22%). The majority (67%) of participants indicated that they had injected drugs at least once, while 35% reported ever experiencing an overdose.
Almost all participants reported experiencing 'problems' related to their drug use. Specifically, the majority (72%) indicated they had experienced 'social problems', followed by 'financial' (60%), 'school-based' (59%), 'health' (45%), 'work' (45%), and 'legal problems' (42%).
Health status. When asked about current health status (reported as 'in the last 30 days') utilizing a 5-point Likert scale (i.e., 1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent) the most common category chosen was 'fair' for both physical (32%) and mental (30%) health status. Notably, a large proportion (24%) reported 'poor' mental health status. In addition, when asked about specific mental health diagnoses, 60% of participants reported ever having received a diagnosis of depression, followed by anxiety (56%), substance use disorder (42%), schizophrenia or psychosis (19%), and a personality disorder (16%); notably, among respondents with a diagnosis of depression, anxiety or substance use disorder (n = 43), 63% reported that they had ever been diagnosed with all three.
# Qualitative results
Treatment or service barriers and deterrents. The participants reported a variety of influential factors that affected their desire or ability to utilize available services or treatments in their communities. These factors were primarily categorized into four main themes: Personal; Social; Structural; and Physical.
Personal barriers or deterrents. One of the major themes that arose when examining barriers to treatment revolved around personal or individual-level obstacles. For instance, many participants indicated that the main reason they had not sought treatment was because they were unaware of which services were available in their community, did not know how to access them, and that there was generally no information about how to navigate the service system.
Regardless of whether or not participants were aware of what treatments or services were available in their community, many youth participants indicated that they did not attend or utilize them, and expressed apathy towards treatment. Some had not thought about getting help, whereas others communicated that they did not want to get help, were not 'ready' to seek treatment, or had a lack of motivation and/or no desire to restrict their drug use. For example, one youth participant responded the following when asked about ever thinking about seeking help for their drug use:
"Not really, because I'm not looking for help right now. Because I'm not ready to quit."
[formula] (Youth Participant 44) [/formula]
Many participants indicated that they wanted to keep using drugs, and that even if they sought treatment or help, or were obliged to go to treatment (e.g., due to court orders, parent's decisions, etc.), it would be futile. Key informants further suggested that many youth have a hard time admitting that they have a problem or need help, and often believe they can handle their problems on their own, especially when it comes to addiction and/or mental health issues.
"Nobody really wants to go get help, like we have a hard time admitting that we need help, especially for things that are like, oh, whatever, I can do it by myself. Like you know, like mental health or like addictions, like nobody wants to admit it."
## (key informant 7)
Social barriers or deterrents. Another common barrier or deterrent for treatment that arose revolved around social barriers. Both the youth and key informant participants stated that there was a lack of confidentiality and privacy in their communities, and that organizations often shared client information amongst each other. Moreover, many youth indicated that they did not want their friends or family to know that they were seeking treatment. Specifically, fear of being judged/stigmatized, letting the important people in their life down, and social isolation were all articulated reservations. As an example, this was expressed by one key informant when they stated:
"They [youth] don't want to acknowledge the family members or repercussions, letting their parents down, whatever way you want to say. Also with their peers or friends they don't want to be the one that says I've got a problem and then get shunned out."
(Key Informant 15) Furthermore, some youth participants and key informants expressed that people are scared of seeking treatment due to the potential consequences of going to the authorities, such as having the police or children's protective services become involved in their lives. For instance, one youth participant expressed this particular concern: ". . .some of my friends, like they have families also, and they're scared to like go forth and make their addictions and problems known to professionals because they're scared to-you know, like Children's Aid might get called or something like that."
## (youth participant 95)
Moreover, many participants noted that their friend groups and/or acquaintances would deter them from seeking help and would sometimes peer-pressure them into skipping appointments or attending treatment. This was specifically the case for those who had to visit treatment centres or services daily (e.g., methadone clinics, pharmacies, needle exchange programs, etc.) as they would often run into acquaintances there. When asked if there was anything that made it difficult for the youth to attend services, one youth participant stated: "Friends . . .like if you hang out with a bunch of people who use and there's drugs right in front of you because they're doing it you're going to want to do your drugs and not go."
(Youth Participant 83) Some youth reported that they did not want to seek treatment because it would be embarrassing or socially challenging. Specifically, stigma, judgment, discrimination and racism were often reported by the participants as reasons for not seeking or continuing treatment. Perceptions of systemic discrimination and stigmatization of drug use were particularly salient, and especially noted as common among certain service providers, such as hospitals. As such, many participants expressed that service provider attitudes and/or lack of professionalism was a deterrent to using services. For example, one youth participant viewed service providers to be judgmental, and stated: "They [service providers] don't understand as much why people do it. They more look at the addiction rather than looking at the reason for the addiction. . .a lot of my friends don't want to go and get help because they feel it's just going to turn into judging, especially because if you go to a hospital or something, you get found with this kind of drug in your system, certain medication they won't give you even if you need it or anything because even if you're a past user, you might. So there's a lot of stereotype and a lot of criticism."
## (youth participant 15)
Structural barriers or deterrents. A third key theme that arose when examining barriers or deterrents to treatment regarded structural barriers. Wait lists to get into treatment or to be connected with a service provider were the most commonly reported deterrents. Being put on a wait list directly affected many participants' motivation to continue to seek treatment. Numerous youth participants expressed that their desire to get help was often fleeting and time-sensitive; if they did not get immediate help when motivated to do so, they would change their minds, forget to attend their appointments, and/or revert back to drug use. "The waiting time . . . I know a lot of people bitch about that. Like they're in the mood to get clean and then by the time the appointment comes they forget it and then it's done. Like they miss the appointment and then they got to do it all over again and then they just say fuck it, let's go get high."
## (youth participant 76)
Other concerns revolved around a general lack of services available in their small communities, and that existent services are extremely under-resourced. Some places, such as emergency shelters, were often too full and/or had limited capacity to take on new clients. For instance, when asked what services or treatments were available for youth in their community, one key informant stated:
"The big thing that I always hear is there's not enough, not enough beds available at either shelters or with treatment, it's just constantly busy. And if somebody needs treatment and they wanted to go in like I feel like they should be able to go in right away."
## (key informant 26)
Other programs (especially detox/withdrawal management and rehabilitation programs) had limitations on the amount of time a client could stay and use their services; typically, once the client was finished their allotted treatment 'period', they were not connected to further service providers.
In addition, limited hours of operation were commonly cited as major barriers to seeking treatment. Most services closed early in the day and/or were only open on certain days of the week. For example, when asked about barriers to accessing services in their community, one youth participant mentioned:
"The Health Unit is closed on the weekends, so that's become a problem before . . .because that's two days in a row that it's closed. . .and they're the only place that provide free, like, supplies."
## (youth participant 21)
Other barriers involved administrative or bureaucratic requirements. Not having proper identification was seen as a major barrier to receiving treatment or accessing/retaining necessary social service assistance such as welfare or disability, which was integral to ensuring a steady enough income to secure safe housing and obtain necessities. Some services required membership fees that the youth did not have the money to cover and could not afford. Furthermore, the overall process of seeking help was seen as overly administratively burdensome and difficult to navigate which would directly deter participants from trying. Other respondents indicated that even if they knew where to go or who to speak to, there was just not enough help and no specific go-to person that could assist them. For example, one youth participant expressed this: "Ontario Works [OW] sure, like welfare is good and all, but it's extremely hard to get on it if you're homeless and you're not in school.
## As a youth, i tried to get on ow after i lost my job, after i had moved out of [service name] that last time, and i tried three times and they wouldn't let me on even though i brought my eviction papers to them, i brought my roe [record of employment] to them. i told them exactly what was going on, i told them exactly why i wasn't going to school. they gave me a list of things to do to qualify for ow, i got all of them done. they gave me 2 days. i still managed to get every single one of them done in those 2 days and they still didn't let me on."
## (youth participant 5)
Physical barriers or deterrents. The final overarching theme that arose when examining barriers or deterrents to treatment focused on physical barriers. In particular, a common concern regarded geographical isolation, as well as a lack of transportation. Many youth participants reported not owning a vehicle and relying on public transportation to get around the city, however, buses ran infrequently, and participants did not have money to afford bus fare so they could not make their appointments. For instance, when asked if there were any problems trying to access services in their community, one youth participant stated:
"Usually transportation if I lived too far from it is always the main concern. . .I take the bus because I can't drive and I don't always have bus fare or money to buy a monthly bus pass."
## (youth participant 4)
Moreover, many youth and key informant participants indicated that due to a lack of services in their home community, the youth often had to travel to other communities to access services. This was viewed as especially difficult due to lack of transportation and the remoteness of certain communities-particularly Northern First Nations reserves-which were difficult to reach. For instance, when asked if their geographic location affected their ability to access services, one youth participant stated: "For me, it's pretty much an isolated reserve, so that's pretty hard. Two times harder than it was in Timmins for me. It's a fly-to area."
## (youth participant 23)
Many youth participants expressed that they had travelled to certain communities to get help (or had been brought there by nature of being involved in the criminal justice system), but once they had finished treatment (or were released from jail), were not able to travel back to the remote communities they came from. This contributed to participants being transient and/or homeless.
In addition, a lack of mobility overall, as well as weather concerns and a lack of weatherappropriate clothing were common issues reported by both youth and key informant participants as major barriers. For example, when asked whether they had personally experienced problems accessing services one youth participant indicated:
## "during the summer no, but during the winter it can be a big problem like i'm not going to walk 25 minutes to a service in blistering cold weather when i don't own a winter jacket or winter boots."
(Youth Participant 18) Treatment or service needs and wants. The youth participants in each community expressed a number of desires for treatments or services that they would need or want to have available. These needs included references for scaling-up services provided by existing organizations, as well as requests for the implementation of specialized and/or new programs and services; these desires were categorized into: harm reduction-based programs; low-threshold programs; specialized and/or group-specific programs; and counselling and/or peer-based programs.
Harm reduction-based programs. The most common request made by both youth and key informant participants when asked what treatment or services they would like to see in their community focused on harm reduction-based programming. For example, many participants suggested a need for a safe injection site or somewhere similar where they could go consume their drugs as well as get their drugs checked or tested to ensure that they were safe.
## "i strongly suggest a safe injection site because we see, on a daily basis, we see needles on a couple of streets and everything now and i wouldn't want my kids stepping on it."
## (youth participant 59)
Scaling-up of needle exchange programs and naloxone provision were also articulated needs, particularly because youth did not have the money to afford buying clean needles from pharmacies. Some participants expressed a need for more places to dispose of their used needles since there were only a few places that had bins to do so in their community, and that this consequently led them to littering their used needles on the ground. Mobile outreach harm-reduction based services which deliver clean needles, necessary health supplies and wound care, as well as dispose of used paraphernalia, were proposed as services that would be particularly helpful. For example, one youth participant suggested:
"Have more outreach vehicles. Even they can go around to the using houses, and collect the dirty ones , and get them clean, and keep it sterile."
[formula] (Youth Participant 1) [/formula]
Lastly, some participants requested general knowledge on safer use practices.
"Somebody that can. . .like a service that shows you how to. . . use drugs safely, like harm reduction."
## (youth participant 78)
Low-threshold programs. The second most common theme that emerged regarding service and treatment needs in the communities referred to low-threshold programs (i.e., programs that do not impose abstinence from drug use and remove bureaucratic or administrative barriers to access). To this end, the most common recommendation by youth and key informants was a youth-specific drop-in centre that was easily accessible. For instance, when asked if there were any services in their community that they would want to have, one youth participant stated:
"Well, first of all, they don't even have drop-in centres. They don't have drop-in centres over here, like somewhere you can go and chill out and not have to pay money."
## (youth participant 26)
Some of the youth participants recognized that there was not a lot to do in their small towns, so they would use drugs as a form of entertainment. As such, other suggestions included a need for increased community-based and/or organized activities. As an example, when asked what would be the most helpful for youth in the community, one participant suggested:
"Maybe like more activities, like sports activities, like free-like some kind of organized activities more often. Like there's nothing around here."
## (youth participant 25)
In addition, flexibility and increased hours of operation-specifically 24-hour availabilitywere commonly suggested. Many youth and key informant participants noted that services are only open limited hours which are not conducive to their lifestyle. One youth participant, like many others, made this need clear when they indicated:
"I think that they should have a 24-hour clinic for people, you know, just in case someone needs a naloxone kit, or just in case someone decides they're going to use a dirty needle, and then they get hepatitis or AIDS. Like they don't know, it could save someone's life. . .like I think that if they had a more flexible sort of an organization that could actually be there when people really actually need it, instead of when it's convenient for them."
## (youth participant 43)
Importantly, a place that offers access to treatment as well as wrap-around services for health and drug use-related issues, in one centralized location, was seen as something that would be most beneficial for youth. This 'centralized' location was seen as crucial to alleviate the burden of needing to travel to different services and in order to mitigate geographical restrictions. One key informant reiterated the need for such a comprehensive and consolidated service:
"Well, the walk-in centre like we were talking about. . .they need more of those. . .also provide like a safe injection site, all that under one roof. One of the problems, like I said, sending people around to different roofs is a bad idea. You need to keep everything in one place."
## (key informant 27)
Specialized and/or group-specific programs. Another common theme that emerged when examining treatment or service needs and wants within the communities revolved around specialized programming. For instance, several youth participants expressed a need for more tailored and youth-oriented programs. Many of the available programs in the communities were either targeted specifically towards adults (some of which would not accept youth), or accepted people from all ages and provided generalized services (which were thus seen as irrelevant or unhelpful for youth). For instance, when asked if there were any services they needed in their community, one youth participant stated: "And we should have a shelter just for young, not mixed with the adults. We should have a shelter here for just the young homeless people like the way they have it in Toronto. They have homeless shelters for the younger ones and the older ones because when the young ones come here they're learning off the older people and that's how they pick up needles and stuff."
(Youth Participant 74) Some of the participants who self-identified as Indigenous suggested that more cultural or traditional-based programming would be beneficial for them as it would allow them to get in touch with their roots and address their issues using the teachings of their culture. Other specific needs included having a safe, open and non-discriminatory space for those who identify as Lesbian, Gay, Bisexual, Transgender, or Queer (LGBTQ) to go and seek treatment that is relevant to the specific issues they were facing. For instance, when asked what services they would want to see in their community, one participant stated: "There should be a place for people that are gay, bisexual, whatever, the LGBT community and stuff like that. Just a separate place for them to go and feel comfortable and not afraid too." (Youth Participant 10)
In addition, cut-off ages where certain programs cannot accept clients under a certain age, or are mandated to remove people from the program once they become a certain age, were seen as unaccommodating. As such, some participants suggested eliminating cut-off requirements and expressed a need for an increase in transitional-based programming to address this issue. Many key informants also re-iterated the need for youth-specific and transitional programming. As an example, one key informant commented on the struggles of helping youth where services are predominately adult-based: "We need . . . mental health services and addiction services targeted specific to youth. . .I'd love to have counsellors that are trained in counselling youth. . .we have no pediatric mental health counsellors. . .I think it really is being in a smaller northern community we don't have the same youth services and it is, it's all about having targeted to youth. I hate having to try and access adult care-whether it's to do with mental health, addictions, LGBT issues-I'm trying to navigate the adult system for kids which I hate. It's not the same, they're not the same people, they need someone who knows kind of the teenage brain and some of the things they're dealing with."
## (key informant 19)
Gender-specific programs were also suggested. This included maternal or pregnancybased-programs that address women's specific needs in a safe and effective manner. For instance, when asked if there was anything they would like to change or improve about the services in their community, one youth participant revealed:
"They need more programs. . .that support pregnant woman and this situation [being addicted to drugs while pregnant]. There should be much more of this kind of programming. . .it's so sad."
## (youth participant 92)
Family-based treatment and services were further suggested as some participants had tumultuous family and home lives, and/or would engage in substance use and criminal activities with their family members. Further, some participants expressed that there were no treatments or services that were specifically tailored to addressing their drug of choice such as cocaine, speed, or crystal meth, and that a program that was drug-specific would be helpful.
"We need something to help everyone get off this crystal meth. There's nothing to help people get off of crystal meth. There's nothing, man. You have to do it cold turkey."
## (youth participant 36)
Counselling and/or peer-based programs. Further needs that arose focused on a desire to have more counselling-based services. Many youth participants articulated that they just wanted "someone to talk to", and someone who was willing to listen to them about the problems they were experiencing. For instance, when asked if there was anything else they needed that they did not have access to in terms of services for their drug use, one youth participant stated: "Probably counselling. Somebody to talk to. Somebody who wants to talk to us."
## (youth participant 29)
In addition, it was noted that services would be more helpful if they were non-discriminatory, and run by people with lived experience or their peers so that the youth would feel more comfortable opening up and relating to someone who has gone through what they have. As an example, when asked what service would be the most beneficial, one key informant mentioned the importance of using peer workers:
"Peer workers. So have a youth as the outreach worker cause who knows a youth better than a youth."
(Key This was also seen as a way for the youth to have their voices heard since they often felt as though they were not included in the decisions that affect them.
# Discussion
The purpose of this study was to examine the barriers, gaps and needs related to treatment and service provision for youth who use illicit and non-medical prescription drugs in Northern Ontario communities. The goal was to solicit valuable input from youth, as well as key informants who work with youth, on how to improve the service system to meet the needs of youth. In addition, we utilized a community-based participatory research approach with an emphasis on connecting with community members, viewing them as equal partners in the research and embedding the research in the specific context of each community; this approach has been found effective, especially when researching marginalized populations.
Findings from this study corroborate many of the conclusions found in existing literature on this topic. For example, remote communities have been found to have critical service gaps for addictions and mental health needs, and Northern and rural communities often have few mental health and addictions services; where these do exist, they are generally fragmented and disconnected from one another. Moreover, an overall lack of services specific to youth addicted to drugs, and a need for prevention, early intervention activities and low-threshold/easily accessible substance use treatments that are youth-oriented has been identified. Other studies suggest that there are common barriers that impact youth and directly affect their decisions, specifically when it comes to seeking treatment such as lack of awareness or motivation, stigma, wait times, paper work, etc..
Although each community is unique and presents their own individual needs, overall, many of the outcomes, sentiments and suggestions were similar across all communities. Specifically, general apathy towards confronting their drug use was common among youth; yet this is not a novel finding since youth have been found less likely to seek help specifically for mental health or substance-use-related issues compared to their older counterparts, and tend to be self-reliant when it comes to dealing with these problems. However, studies have found that among youth, perceived benefits of treatment can predict help-seeking behavior and can outweigh perceived barriers. As such, increasing motivation to seek treatment among youth, and promoting and reinforcing the benefits of treatment in a way that is most likely to resonate with them is necessary in order to increase its appeal and uptake, and should be a focus when it comes to treating youth with substance use problems. In order to achieve this, it is imperative to have open discussions with youth about the benefits of seeking treatment and speaking to someone about their problems, especially since adolescence is a critical time point for physical and psychological development, and is a high-risk period for drug use initiation.
Nonetheless, even when youth become motivated to seek treatment and understand its benefits, stigmatization remains a major deterrent. Fear of stigmatization and beliefs about the lack of helpfulness of treatments and services for addiction and mental health issues has been correlated with help-seeking intentions among youth, and many studies have reported this as a specific barrier for treatment. Drug use is inherently stigmatized, and particularly so among youth; studies have shown that stigmatization experienced in youth can carry into adulthood and can affect personal drug use experiences and trajectories. As such, it is important to tackle stigmatization and offer services that youth perceive as socially inclusive and approachable. One way to facilitate this is to offer services run by non-discriminatory and empathetic counsellors and/or peers, and to recognize that lived experience provides unique contributions towards positive change and ultimately, the reduction of stigma.
When it comes to treatment and service needs, low-threshold and/or harm reductionbased services were the most commonly reported need. Low-threshold programs have been defined differently, but essentially include 'inviting' atmospheres, effective client engagement, confidential service delivery, tailored services, and peer support to reduce stated barriers to service access. As a primary example, many youth participants suggested drop-in centers with on-the-spot help available 24 hours a day, run by peer counsellors as a service they would use and benefit from. Notably, there have been a number of low-threshold harm reductionbased interventions implemented in some of these communities since our data was collected. For example, supervised consumption and/or overdose prevention sites (both legally sanctioned and non-legally sanctioned) as well as rapid access addiction medicine clinics (i.e., lowbarrier, walk-in clinics where patients can go to get help for a substance use disorder without referrals) have been executed in a number of Northern Ontario communities such as Thunder Bay and Sudbury.
Moreover, there is an overall low-barrier 'integrated youth services' movement unfolding in Ontario (and Canada more broadly), which includes a number of initiatives and programs that provide easy access to brief interventions, peer support, care navigation, primary care and a point of access to higher intensity services for both mental health and addictions issues for youth. For example, in Ontario specifically, 'youth wellness hubs' have now been implemented in a number of communities across Ontario (including in the North), which were established to serve as integrated 'one-stop-shops' for youth aged 12-25 who are dealing with mental health and addictions issues; these include many of the services that youth suggested would be beneficial to them (such as peer services, outreach, system navigation, etc.). Therefore, since our data collection concluded, there have been a number of youthoriented interventions that have been implemented which are slowly expanding and ideally addressing many of the issues that youth reported experiencing.
Although these are positive efforts, many have faced strong political opposition which has hindered their effectiveness, and they have not been implemented equally across all communities (e.g., there are currently only 10 'youth wellness hubs' across all of Ontario). As such, more needs to be done to address the youths' articulated needs.
Additional research on this important topic is highly needed. Studies that geographically and/or spatially map all treatment and service providers available in Northern Ontario would be helpful to gaining a better understanding of the availability of these services. Moreover, quantitative studies examining treatment utilization and transition of care within each community using publically-available health record data and databases (e.g., Ontario Health Insurance Plan (OHIP); Institute for Clinical Evaluative Sciences (ICES), etc.) could help better inform whether or not and how often youth are seeking help for drug use in their communities. Of note, involving people with lived and living experience, as well as Indigenous people and culturally-specific organizations, throughout the entirety of such studies (including in conceptualization, design, implementation and analysis) is crucial to ensuring that the results truly reflect the experiences of the participants and that the knowledge created is accessible to those that can benefit from it.
The results of this study reflect the perceived experiences of the participants in their unique community-based settings, and were affirmed by key informants during the feedback process. However, the results should be interpreted with caution and limitations to this research must be noted. This study was a mixed-methods study, and as such, is susceptible to biases inherent in self-report data such as memory or recall bias, response bias, interpretation of the questions, and social desirability; this may be particularly true since focus groups have a tendency to elicit a dominant voice, and the potential for posturing for the microphone exists. Further, some participants may have lied about their age due to the incentive of the participant honorarium, but measures to mitigate this possibility were taken. Moreover, given that drug use is heavily stigmatized, particularly in smaller communities, participants may have under-reported dimensions of their drug use and any discrimination or prejudice experienced. The information gleaned from this study is also not generalizable for a number of reasons: the study only reports on the 11 Northern Ontario communities visited, so the findings are specific to youth who reside in these communities; the number of participants recruited from each community widely varied due to a number of factors, e.g., lack of eligible participants, weather conditions, level of engagement with community partners, number of services available in the community, etc.; the sample was purposeful, and many of the youth had pre-existing relationships with the service providers in the community and may therefore under-represent those who have no connection to the service system, although we utilized a respondent-driven sampling technique in order to mitigate this bias. Moreover, the mobile lab was parked in pre-determined places where it would be the most visible to our target population, and potential participants may have felt insecure or uncomfortable to be seen in the mobile lab. Further, we may not have reached potential participants who live in more remote areas and cannot travel or reach the mobile lab due to a lack of transportation, which was pinpointed as a major barrier to accessing services.
# Conclusion
The findings from this study suggest that youth who use illicit and non-medical prescription drugs in Northern Ontario communities would benefit from community-based implementation of low-threshold services such as a drop-in centers/emergency shelters tailored specifically to youth-available and open around the clock-which provide education about the benefits of treatment as well as service system navigation, activities, and wrap-around services. These services should mandatorily provide rapid-access medical assessment/treatment, counselling and programs run by peers and non-judgmental and empathetic professionals in order to be optimally effective for youth. In addition, it would benefit both youth as well as community service providers to implement a system or arrangement (e.g., monthly community consultations) that includes youth in the decision making process and which improves integration and collaboration across organizations and sectors within each community. This would allow for a deeper understanding of the ongoing issues and needs of youth in each community, and provide for better community connections to address these needs.
This study helps to better understand the needs of the youth who reside in Northern Ontario communities, which is essential since it was conveyed that many of the currently available services are not relevant or effective for them. Moreover, although our findings may not be generalizable, the knowledge gained is relevant and can be applied to many remote and rural community settings across Canada (and North America more broadly). Thus, the presented results have the potential to aid in evidence-based decision making for the development and implementation of future services and interventions for youth who use illicit drugs which actually reach the youth, address their articulated needs, and help them deal with their substance use issues.
## Supporting information
|
The Effectiveness of Psychosocial Interventions for Elder Abuse in Community Settings: A Systematic Review and Meta-Analysis
As a global public health concern, elder abuse negatively affects health, psychosocial wellbeing, and mortality among elders. Research and practice efforts made to explore effective prevention and intervention strategies are growing. Despite the growing number of intervention studies on elder abuse, research synthesis on the empirical literature seems lacking. This study aims to identify the pooled effect size of prevention and interventions targeted ultimate and intermediate outcomes for elder abuse that occurred in community settings. Following the Cochrane guideline, our team searched across eight electronic databases and manually searched reference lists of eligible studies and existing systematic reviews for all potentially eligible studies. A random-effects model of 51 effect size estimates reported an overall positive and statistically significant treatment effect of psychosocial interventions for elder abuse, d = 0.63, p < 0.05. The overall treatment effect was approaching statistical significance at 0.1 level for ultimate outcomes, d = 0.32, p = 0.09, and intermediate outcomes, d = 0.75, p = 0.1. An overall significant effect size was found among family-based interventions, d = 0.59, p < 0.05, and interventions targeting older adults and their caregivers, d = 0.45, p < 0.05. Existing evidence supports an overall significant effect for psychosocial interventions for elder abuse. Interventions that used a family-based model, combined education and supportive services, and targeted both caregivers and elders, showed significant effect size, suggesting such features being considered in elder abuse intervention design. Future intervention research is needed to shed light on the link between intervention activities and ultimate change in elder abuse behaviors.
# Introduction
Elder abuse refers to intentional or unintentional harmful acts toward an older person where trust is expected. Common types of elder abuse include physical, psychological, sexual abuse, financial exploitation, and neglect . As a global public health concern, about 1 in 6 community-dwelling older adults experienced some form of abuse in the last 12 months as found in a meta-analysis that explored prevalence rates of elder abuse. The prevalence rate of abuse in elder care facilities is even higher as two-thirds of care staff members reported abusive behaviors. The profound impact of elder abuse on victim's health, finances, quality of life, and even mortalitydeserves attention.
Current elder abuse interventions include public education and advocacy, caregiver support, psychological support for victims, care coordination, and multidisciplinary case management to name a few. Based on existing literature, programs that suggest the effectiveness of elder abuse are through providing (1) caregiver supportive services, (2) money management coaching, (3) telephone helplines, (4) emergency shelters for older victims, and (5) access to a multidisciplinary team. Educational or training interventions are more accessible than supportive services or case management interventions. Elder abuse preventions and interventions tend to achieve two types of outcomes: reducing the occurrence of abusive behaviors, and mitigating elder abuse risk factors, such as psychosocial stress and lack of awareness or competency among nursing staff, family caregivers, and older adults themselves.
Most meta-analysis studies regarding elder abuse focus on the prevalence and risk factors rather than on elder abuse intervention effects. Despite the growing number of intervention studies on elder abuse, research synthesis on the pooled effect for elder abuse interventions remains lacking. From the limited available systematic reviews of elder abuse intervention programs, either found a lack of sufficient evidence to establish the effectiveness of elder abuseprograms or programs that did not use rigorous evaluations to assess the effectiveness of said programs.
Furthermore, the only meta-analysis to our knowledge synthesized 24 studies of elder abuse interventions. This study found an overall small but significant treatment effect for restraint use reduction, d = −0.24, 95% CI (−0.38, −0.09). While highly valuable, this meta-analysis primarily focused on the restraint use as an abuse outcome, which can be questionable as the use of physical restraints with a physician's order can be medically necessary rather than abusive.
Despite the above evidence that supports the potential effect of intervention programs for elder abuse, the conclusion is far from definitive because of the limited number of studies and a narrowed scope of outcomes reviewed in these systematic reviews. Elder abuse preventions and interventions typically aim for changes in intermediate or ultimate outcomes. Ultimate outcomes refer to the reduction in the occurrence or reoccurrence of abusive behaviors, while intermediate outcomes include the mitigation of risk factors (e.g., psychosocial stress) and promotion of protective factors (e.g., improving knowledge of abuse and enhancing the competency of addressing abuse) that will lead to the reduction of elder abuse. Both intermediate and ultimate outcomes need to be considered for a competent narrative for the effectiveness of elder abuse interventions.
Therefore, our systematic review and meta-analysis study aims to examine prevention and intervention studies that targeted ultimate and intermediate outcomes for elder abuse that occurred in community settings. Acknowledging that these two outcome types of elder abuse, though interrelated, reflect distinctive features, we hence examined the pooled effect of all outcomes combined and the respective effect for different outcome types.
# Materials and methods
## Search strategy
The identification of relevant studies was performed in two steps. The first step consisted of searching eight academic databases from Jan 1990 to December 2020: Cumulative Index to Nursing and Allied Health (CINAHL), Cochrane Library/ Central Register of Controlled Trials, Database of Abstracts of Reviews of Effects (DARE), PsycINFO, PubMed, MEDLINE, BIOSIS, and Science Direct. Key search terms included "older abuse, " "elder abuse, " "elder neglect, " "elder mistreatment/ maltreatment, " or "older neglect" and "physical abuse, " or "emotional abuse" or "financial abuse" and "intervention" or "program. " To identify the right design of the intervention, search terms included "control group" or "RCT. " The set of keywords was used for both study title and study abstract search across databases. Second, the reference lists of studies and systematic reviews identified from the database search were reviewed for additional relevant studies.
## Inclusion criteria
The inclusion criteria for studies published in English are those that (1) assessed the effectiveness of a prevention or intervention program designed to address abuse or neglect of elders aged 65 or older living the community settings, (2) targeted at elders or family members, (3) used randomized controlled trials (RCTs) or controlled trials (without random assignment), (4) reported at least one elder abuse intervention outcome, and (5) reported statistical information sufficient to calculate effect size required for meta-analysis. We did not limit the location of the studies. Outcomes consisted of (1) ultimate outcomes: occurrence or reoccurrence of elder abuse behaviors; and (2) intermediate outcomes: reduction of risk factors for elder abuse, such as reduced stress, improved knowledge of abuse, and enhanced competency of addressing abuse.
## Data collection and extraction
Three researchers conducted data collection and extraction. Two reviewers (YS and FS) screened titles and abstracts for eligible studies independently with decisions blind to one another. If any disagreement existed, two reviewers would discuss first and if unresolvable, a third reviewer (AZ) would intervene and make a final decision. The three-person team adopted the same process in later full article review and quality assessment. Inter-screener reliability was 92% and inter-rater reliability of full articles was 85%.
Measures of ultimate outcomes included occurrence or reoccurrence of elder abuse, here termed as behavioral outcomes, and intermediate outcomes (i.e., risk or protective factors for elder abuse) included psychosocial stress, knowledge of elder abuse, and competency of addressing elder abuse.
Basic information extracted from studies consisted of participant demographic characteristics (i.e., age, gender, and ethnicity), geographic areas of the study (i.e., United States, Asia, and Europe), and study design (i.e., RCT and non-randomized controlled trial). Intervention characteristics consisted of intervention type (i.e., family, individual, and mixed), target population (i.e., elders, family caregivers, or both), intervention approach (i.e., education, supportive services, and mixed), and intervention frequency and duration.
## Risk of biases assessment
Quality of studies was assessed using the Jadad scale, also known as the Oxford quality scoring system. Studies' risk of bias was assessed using the Cochrane Collaboration's tool for risk of bias in randomized trials. The research team resolved discrepancies and reached a consensus on these ratings. Publication bias was assessed using a funnel plot (visual analysis) and the Vevea and Wood sensitivity weighted function analysis (statistical analysis;. For funnel plot, we plotted independent effect sizes only first and then plotted all effect sizes with some of them are dependent of each other.
## Meta-analytic procedures
We used the R software for data analysis. Treatment effect sizes were estimated for each individual study to determine treatment clinical effects. For continuous outcomes, the standardized mean difference (SMD) was calculated to obtain Hedges' g statistic. For binary outcomes, an odds ratio (OR) was calculated first, followed by taking the log transformation of the odds ratio (i.e., log odds ratio). The log OR statistic was further transformed into the same effect size metric as the Hedges' g statistic using procedures suggested by. The Hedges' g was further bias corrected using a J functionto obtain an unbiased estimation of the treatment effect, noted as d for the rest of the text. When meta-analyzing the effect size estimates, we used the inverse variance weight, which is considered as an optimal weight estimate in meta-analysis (Marín-Martínez and Sánchez-Meca, 2010).
Between-study and between-effect size, heterogeneity was assessed using multilevel modeling with R's metafor package. A pooled overall treatment effect and potential moderator analyses were achieved through metaregression with robust variance estimation (RVE) using R's robumeta package. The intercept only in the meta-regression model offered overall averages of treatment effect sizes across studies; and models with covariates allowed the identification of effects of potential moderators on treatment effect sizes. Meta-regression using the RVE method effectively handles the statistical dependence created by one study reporting multiple effect size estimates on the same outcome. For example, a study may use more than one measure to evaluate a provider's knowledge of elder abuse, resulting in the two knowledge measures within the same study being potentially dependent on each other (see, outcome measures). The RVE approach not only effectively addresses the dependent issues but also produces robust estimation regardless of the heterogeneity assumption, meaning results robust across fixed-and random-effects models. Given the small number of studies included in this review, we also conducted a small sample size correction to the meta-regression analysis, and for an estimate with degrees of freedom greater than 4, p < 0.05 is considered statistically significant. For an estimate with degrees of freedom lower than 4, p < 0.01 is considered statistically significant. Sensitivity analyses were conducted by averaging dependent effect sizes within each study. Because both methods produced the same statistical inference, we reported the RVE results in this paper.
# Results
# Search results
The review process is summarized in. A total of 2,986 studies were identified through a comprehensive search strategy for interventions to prevent or stop elder abuse in the community or an institutional setting. After 1,578 duplicate studies were removed, 1,255 studies were further excluded based on a title and abstract review. Of the remaining 153 studies, 147 studies were excluded for reasons, such as single arm trial without a control group or without statistical data, resulting in an analytical sample of six studies, containing 51 effect size estimates, in the final meta-analysis.
## Characteristics of included studies
Study characteristics are presented in. Of the six primary studies, four studies (67%) were published after 2010 and one (17%) was published in the 1990s. The six studies included a total sample of 1,305 participants. Participants' ages averaged at 64.65 (SD = 4.15), and 64.52% were female (SD = 17.13). Four studies (66.7%) included at least one intermediate outcome measure and four studies (66.7%) included at least one ultimate outcome on elder abuse. Specifically, three studies included psychosocial stress outcomes, four studies encompassed outcomes related to knowledge and competency, and four studies addressed ultimate abuse-related behavioral outcomes.
Most interventions in reviewed studies were delivered in a family format (n = 3, 50%), which lasted on average over 10 months (SD = 7.55, range = 3-18 months). Two individualbased interventions (33.3%) were on average 21 months in duration (SD = 4.24, range = 18-24 months) across studies. Three studies directed at both older adults and family members, two studies targeted older adults and one study targeted family members. Two interventions adopted an education approach, three used supportive services, and one incorporated both educational and supportive service interventions. Study sites covered Europe (n = 2, 33.3%), the United States (n = 2, 33.3%), and Asia (n = 2, 33.3%).
## Quality assessment of included studies and risk of bias
Both RCT studies (n = 4) and non-randomized controlled trial studies (n = 2) were rated using the Jadad scalefor reporting controlled trials and the Cochrane risk of bias assessment tool. Using the Jadad scale (see, the six trials had an average score of 2.8 (SD = 1.17) out of 5, indicating acceptable quality. These studies were rated satisfactory in mentioning randomization (5/6), tracking all participants (6/6), and randomization (4/6). However, these studies done were not satisfactory using appropriate blinding, just two studies mentioned blinding. Using the Cochrane Collaboration's tool for assessing the risk of bias (see, studies were rated most satisfactorily in selective outcome reporting (6/6), random sequence generation (4/6), and handling incomplete outcome data (4/6). Risk of bias was observed in allocation concealment (0/6), blinding of study participants and personnel (0/6), and blinding of outcome data assessment (2/6).
Both visual and statistical examination suggested an absence of publication bias (see. The funnel plot seemed reasonably symmetric and showed no concerning outliers. The Vevea and Wood sensitivity weighted function analysis further confirmed the absence of publication bias.
## Meta-analysis results and subgroup analyses
Multilevel modeling and the Q statistic used in heterogeneity assessment indicated a significant amount of heterogeneity across both effect size estimates and studies [Q(50) = 5914.3, p < 0.01], suggesting a random-effects model is appropriate. We found an overall positive and statistically significant treatment effect (d = 0.63, 95% CI [0.02, 1.24], p < 0.05).
Effect sizes were found depending on outcomes in subgroup analyses. The overall treatment effect was approaching statistical significance at 0. In terms of intervention characteristics, we found that significant effect sizes across intervention type and target population. An overall significant effect size was found among family-based interventions (d = 0.59, 95% CI [0.18, 1.01],
## Moderator analyses
Results of moderator analyses are presented in. Only two univariate meta-regression models revealed a significant moderator that explains the heterogeneity between studies and effect sizes. Interventions targeting older adults reported significantly greater treatment effects than those targeting caregivers, or both. Intervention using a family centered approach had a greater treatment effect than the intervention delivered using a mixed format which included group and individual interventions.
# Discussion
Our major finding identified a statistically significant effect size (d = 0.63) of psychosocial interventions for elder abuse, suggesting that the effectiveness of available elder abuse prevention and interventions are moderately supported by evidence. Addressing elder abuse needs a multisystem effort that targets various change agents (e.g., elders, family, and caregivers) and along with multiple domains such as abuse awareness, knowledge, and behaviors . Although the overall effect size sounds promising, the findings from the subgroup analyses shed further light on the effectiveness regarding different intervention outcomes and provided specific directions for future research. The treatment effect for intermediate outcomes (d = 0.75, p = 0.1) and for the ultimate outcomes (d = 0.19, p = 0.09) approached significance at 0.1 level. Because of the small power due to a limited number of articles included in the metaanalysis, these findings, though should be interpreted with caution, appear encouraging and promising. Intermediate outcomes can be viewed as mediators which bring about elder abuse reduction. Interventions reviewed in this study leaned toward an effective impact on intermediate outcomes, such as psychosocial stress, knowledge, and competency. This finding corresponds to a conclusion based upon an earlier Cochrane review of interventions for abuse prevention. As abuse-related knowledge outcomes are often treated as intermediate outcomes, research needs to improve their knowledge-related outcome measures to capture notable changes in abuse awareness and attitude, along with their impact on the occurrence of elder abuse. Moreover, research should not stop arriving at satisfactory intermediate outcomes, but rather continue to disclose the pathway between changes in psychosocial stress and abuse reduction.
Effect size is significant among interventions that were family-based, geared toward elders, and family members, and adopted multiple approaches (e.g., combining education and services). The effectiveness of family-based interventions suggests the change of family dynamics can mitigate abuse. Consistently, an overall statistically significant treatment effect was observed among interventions targeting both elders and family members. As we had a limited number of interventions solely on elders or caregivers, we could not speculate the effect size for each respective group. Elder abuse mostly occurs in family settings, where perpetrators are most family members. Thus, targeting family issues through mitigating risk factors on both sides, potential victims and perpetrators, are likely to yield ideal outcomes.
The use of various intervention approaches (e.g., education and supportive services) appeared to be effective, as mixed approach interventions showed an overall significant treatment effect. As there are multiple risks for elder abuse, intervention using mixed approaches can be more effective than interventions using one approach. Future interventions need to incorporate education, support, and services to assist older victims. Enhancing one's awareness of elder abuse through education is important to the prevention of abuse; however, information itself is far from sufficient to trigger changes in attitude and behavior. Often, elders vulnerable to abuse or have been abused are likely to have other pressing social, health, financial, and legal needs. Similarly, mixed approaches better serve family caregivers who need tangible support to help manage challenging care tasks and emotional support for stress and burden.
Several limitations are inherent to systematic review and meta-analysis studies that should be noted. First, chances could be that one or more studies were missed in our search.
The total number of studies included was small, and a limited sample in subgroup analyses may have failed to show significance due to insufficient power. The small sample size, i.e., a small number of included studies, also prevented us from conducting certain subgroup analyses such as an overall treatment effect for interventions targeting different target population. Future studies should consider these analyses when more studies become available. Second, while two coders independently coded all studies and a team of experienced researchers resolved any disagreements, the results of this study are still subject to human errors. Third, as we used an advanced method to synthesize effect size estimates across outcomes, individual outcome contains distinctiveness in conceptualization. In the subgroup outcome analyses, some outcomes may suffer from a small sample size. As a result, we were unable to identify if the non-significant findings were due to low power. The same issue may exist in the moderator analyses, which prevented us from drawing any definitive conclusions. Fourth, as we focused on our search for psychological interventions, we did not include legal or policy interventions. As this area may play a role in elder abuse, these should be considered in future studies. Furthermore, research with a larger number of studies included will help explain the variations across effect sizes and studies.
# Conclusion
This study represents an initial effort to examine the pooled effect of elder abuse preventions and interventions via metaanalyses. Existing evidence is supportive of a modest effect (approaching significance at 0.1 level) of psychosocial interventions for elder abuse. Evidence appears promising for interventions on modifiable intermediate outcomes such as psychosocial stress, knowledge, and competency that are typically theorized to lead to changes in elder abuse occurrence, as well as interventions targeting ultimate outcomes (i.e., abuse reduction). Interventions that used a family-based model, combined education and supportive services, and targeted both caregivers and elders, showed significant effect size, suggesting such features incorporated in elder abuse intervention design. Yet, more research evidence is still needed, in particular, research that will shed further light on the link between intervention activities and changes in elder abuse behaviors.
Community-based approach that draws on the concerted efforts from multiple stakeholders (e.g., elders, families, and service professionals) and tackles multi-domain elder abuse risk factors (e.g., knowledge, competency, and support) is worth pursuing. Geriatric health and social care providers as well public health workers in the field of aging should be updated on the status of currently available interventions, adapt evidenced-informed interventions, and account for the heterogeneity of factors at the individual, agency, and cultural levels when promoting a safe and free of abuse environment for older adults.
# Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
# Author contributions
YS screened and coded all studies, drafted the entire manuscript. FS screened and coded all studies, drafted the discussion section of the manuscript, and proofread the entire manuscript. AZ designed the search strategy, resolved conflict, conducted preliminary analysis, drafted the entire manuscript, and proofread the entire manuscript. KW conducted all the meta-analysis and led the writing of the method and result section and proofread the entire manuscript. All authors contributed to the article and approved the submitted version. |
Iatrogenic perforation of a pulmonary artery side branch—a case report
# Introduction
Current guidelines recommend that right heart catheterization (RHC) be considered part of the diagnosis and management of patients with known heart failure who remain severely symptomatic despite initial standard therapies and whose haemodynamic status is unclear. [bib_ref] ESC Guidelines for the diagnosis and treatment of acute and chronic heart..., Ponikowski [/bib_ref] Further, RHC is recommended for the work-up of chronic thrombo-embolic pulmonary hypertension to confirm the diagnosis and assess the severity of haemodynamic impairment. [bib_ref] ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: the Joint..., Galie [/bib_ref] Learning points - Pulmonary artery rupture is considered the most dangerous complication of a right heart catheterization.
- Rapid action and an interdisciplinary approach are one of the pillars in the treatment of this complication.
- Endovascular coil embolization may be considered as a safe and quick procedure when balloon occlusion fails.
This article was judged to be one of the best cases presented at the German Cardiac Society Annual Congress 2020. It was reviewed by team from that organization. * Corresponding author. Tel: þ49651208981911, Email: [email protected] Handling Editor: Amardeep Ghosh Dastidar Compliance Editor: Linh Ngo Supplementary Material Editor: Nida Ahmed V C The Author(s) 2021. Published by Oxford University Press on behalf of the European Society of Cardiology. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] European Heart Journal -Case Reports (2021) 5(6), However, due to its invasiveness, RHC is linked to a risk of complications, ranging from access-site-associated complications and arrhythmias to life-threatening complications, such as injury to the right heart or pulmonary arteries. In a systematic review of 7218 RHC procedures, Hoepker et al. [bib_ref] Complications of right heart catheterization procedures in patients with pulmonary hypertension in..., Hoeper [/bib_ref] observed four fatal events that resulted in an overall procedure-related mortality rate of 0.055% (95% confidence interval 0.01-0.099%). Pulmonary artery rupture is considered the most dangerous complication, with an estimated incidence of 0.031% and 0.05% and mortality rates of 70% and 50%, respectively, in two studies with large sample sizes. [bib_ref] Pulmonary artery rupture associated with the Swan-Ganz catheter, Kearney [/bib_ref] [bib_ref] Swan-Ganz catheter-induced severe complications in cardiac surgery: right ventricular perforation, knotting, and..., Bossert [/bib_ref] Overall, despite the low rate of life-threatening complications, the literature lacks clear recommendations regarding their management, often necessitating individual decisions about a patient's further care.
## Timeline
## Case presentation
We report an 80-year-old female (body mass index 20.8 kg/m 2 ) who presented to our clinic for further evaluation of her dyspnoea [New York Heart Association (NYHA) functional classification III]. Her cardiovascular risk factors included arterial hypertension, and her relevant medical history contained atrial fibrillation, rheumatoid arthritis, and a bilateral pulmonary embolism in 2014 after varicectomy. The patient was on diuretics, a beta-blocker, an angiotensin-converting enzyme inhibitor, and oral anticoagulation.
The physical examination revealed the following abnormalities: mild oedema of the lower legs bilaterally and reduced breathing sounds bilaterally. Auscultation of the heart showed a holosystolic murmur (II/VI) at the apex and left lower sternal border. The echocardiography demonstrated normal left ventricular function and moderate mitral and tricuspid regurgitation. The right ventricle was enlarged, with preserved function (tricuspid annular plane systolic excursion: 21 mm). In the continuous wave Doppler measurement, pulmonary artery systolic blood pressure was significantly elevated at 86 mmHg. Coronary heart disease and pulmonary venous congestion had been ruled out by coronary angiography and chest X-ray, respectively. However, there were small pleural effusions on both sides.
Due to the patient's complaints and the pathological findings, we decided to examine the right heart with a balloon-tipped pulmonary artery catheter (PAC) for further clarification. The RHC was performed in the catheter lab via the right femoral vein. We used a standard balloon-tipped PAC (5French, CORODYN TM , B. BRAUN), which was placed in the left pulmonary artery under radiological control. Immediately after the pulmonary capillary wedge pressure measurements, haemoptysis occurred, which rapidly increased in intensity and led to respiratory failure. This event was accompanied by rapid haemodynamic deterioration, necessitating cardiopulmonary resuscitation (CPR) for 4 min. The patient was orally intubated and ventilated immediately.
Concurrently, angiographic imaging of the left pulmonary artery was performed via a 6 Fr multipurpose guiding catheter (MB1, Medtronic). As a result, rupture of a lateral branch of the left pulmonary artery could be identified as the cause of the haemorrhagic shock [fig_ref] Figure 1: Contrast agent extravasation following perforation of a pulmonary artery side branch [/fig_ref]. Due to the patient's comorbidities and age, we initially decided against surgical treatment, in consultation with the heart surgeon. We inserted a coronary intervention wire (Runthrough V R , TERUMO V R ) into the periphery of the perforated vessel and performed balloon occlusion [Maverick Monorail TM , Boston Scientific; 4.5 mm  20 mm; pressure: 6 atmosphere (atm)], proximal to the perforation. The new angiographic image now showed a complete balloon occlusion of the perforated vessel [fig_ref] Figure 1: Contrast agent extravasation following perforation of a pulmonary artery side branch [/fig_ref]. Balloon occlusion was continued for a total of 40 min.
Meanwhile, the patient's haemodynamic situation stabilized. The activated clotting time (ACT) was 123 s. The repeat angiographic image continued to show contrast agent leakage from the perforation site after 40 min. Due to persistent bleeding from the perforated pulmonary artery, we decided to embolize the perforated vessel with microcoils (Hilal Embolization Microcoils TM , COOK V R , Medical). For safety reasons, to control bleeding and ensure a stable haemodynamic situation throughout the planned bail-out procedure, we decided to implant the coils through a second interventional catheter. To this end, another 6 Fr MB 1 guiding catheter was placed in the left pulmonary artery via the left femoral vein. We then used this catheter to insert a second wire (Runthrough V R , TERUMO V R ) into the target vessel past the blocked balloon. For this purpose, the inflation pressure was briefly reduced to 2-4 atm. We then advanced a microcatheter (Finecross V R MG, TERUMO V R ) via the second interventional wire. The perforation cavity was selectively visualized via the inserted microcatheter [fig_ref] Figure 2: Selective visualization of the perforation cavity via the inserted microcatheter [/fig_ref] , after which embolization was performed with four coils into the cavity. The angiographic image showed complete embolization of the perforation cavity [fig_ref] Figure 2: Selective visualization of the perforation cavity via the inserted microcatheter [/fig_ref]. Consequently, the patient's haemodynamic situation stabilized at this point. Following the procedure, the patient was transferred to the intensive care unit while haemodynamically stable. Sonography revealed a haematothorax on the left side, which was relieved by drainage. Laboratory findings showed a decrease in haemoglobin concentration from 12 g/dL to 9 g/dL. The patient was extubated on the following day. During her stay, she developed pneumonia, which was treated with antibiotics (ampicillin/sulbactam). Ten days after the index procedure, a computed tomography scan of the thorax was performed, which showed a favourable result after coiling [fig_ref] Figure 3: Computed tomography scan showing coils in the left pulmonary artery [/fig_ref]. In addition, there was no evidence of chronic thrombo-embolic pulmonary hypertension.
Due to the patient's pre-existing atrial fibrillation, with significant atrioventricular block in the clinical follow-up, a permanent pacemaker was implanted before discharge. On Day 26 post-index event, the patient was discharged without further sequelae. The echocardiography at discharge showed a systolic pulmonary artery pressure of At follow-up, 1 month later, the patient reported no further episodes of haemoptysis or any discomfort related to the pulmonary artery rupture. Echocardiographic findings were without significant changes.
# Discussion
Right heart catheterization is an invasive method for further evaluating heart failure in patients whose haemodynamic status is unclear and in the continuing diagnosis of pulmonary hypertension. 1,2 Serious complications of RHC occur in 0.1% of cases and pulmonary artery rupture is considered the most dangerous complication. [bib_ref] Pulmonary artery rupture associated with the Swan-Ganz catheter, Kearney [/bib_ref] [bib_ref] Swan-Ganz catheter-induced severe complications in cardiac surgery: right ventricular perforation, knotting, and..., Bossert [/bib_ref] The following risk factors are associated with an increased risk of pulmonary artery injury in PAC: age > 60 years, female gender, pulmonary hypertension, systemic anticoagulation, long-term steroid use, and surgically induced hypothermia. [bib_ref] Pulmonary artery pseudoaneurysm after Swan-Ganz catheterization: a case presentation and review of..., Nellaiyappan [/bib_ref] Our case had four of these risk factors.
Treatment options for pulmonary artery rupture include surgery and endovascular therapy. Due to the rapid deterioration in haemodynamic and respiratory function, rapid diagnosis, and initiation of therapy are essential for survival. In parallel with the interventional treatment below, additional measures should be taken simultaneously, such as activating the CPR team, the anaesthesiologist and cardiac surgeon; resuscitation, if necessary; administration of intravenous fluids in case of hypotension; securing the airway with a selective lung intubation; and protamine injection if heparin was administered.
There are no clear recommendations for interventional treatment of this rare but life-threatening complication. Existing recommendations are based on case reports, 7-11 including selective thrombin injection into the vessel, percutaneous embolization with an AMPLATZER TM vascular plug (Abbott), implantation of a covered stent and coil embolization. [bib_ref] How should I treat a pulmonary artery rupture occurring during a right..., Boukantar [/bib_ref] [bib_ref] Bail-out technique for pulmonary artery rupture with a covered stent in balloon..., Ejiri [/bib_ref] [bib_ref] Repair of a perforated pulmonary artery due to a Swan-Ganz catheter using..., Dobies [/bib_ref] [bib_ref] Endovascular treatment of a ruptured pulmonary artery aneurysm in a patient with..., Ianniello [/bib_ref] [bib_ref] Coil embolization treatment in pulmonary artery branch rupture during Swan-Ganz catheterization, Gottwalles [/bib_ref] In our case, we decided on a step-by-step approach with balloon occlusion first to stabilize her haemodynamics. After checking for adequate ACT and failure to stop bleeding after prolonged balloon occlusion, the heart team reconvened to hold an interdisciplinary discussion. The risk of emergency surgery was considered to be unacceptably high. Although implantation of a covered stent can maintain perfusion of the treated artery, 7,10 we decided against this strategy, because the distal landing zone of the stent remained unknown, despite angiographic controls. Similarly, implantation of a vascular plug was not an option in this case, because the angiographic controls did not allow for an accurate assessment of the anatomy that was distal to the perforation. The simultaneous insertion of a microcatheter past the blocked balloon permitted us to selectively visualize the perforation cavity and subsequent coil implantation without having to compromise the gain in stability. This method is a safe and rapid approach to pulmonary artery perforation when balloon occlusion fails and when the patient is considered to be inoperable or at high risk. A disadvantage of this method is that a local pulmonary infarction is induced. However, this drawback must be considered after weighing the risk-benefit ratio.
# Conclusion
Perforation of a pulmonary artery during pulmonary artery catheterization is an acute threat for which treatment options are limited. Rapid action and an interdisciplinary approach are the pillars in the treatment of this complication. Endovascular coil embolization may be considered a safe and quickprocedure when balloon occlusion fails.
## Lead author biography
Dr Juergen Leick completed his medical studies at the Justus Liebig University in Gießen (Germany). He received his cardiology training at the Kerckhoff Clinic in Bad Nauheim. He currently works as a senior physician in the field of interventional cardiology at the Heart Centre Trier.
# Supplementary material
Supplementary material is available at European Heart Journal -Case Reports online.
Slide sets: A fully edited slide set detailing these cases and suitable for local presentation is available online as Supplementary data.
## Consent:
The authors confirm that written consent for submission and publication of this case report including images and associated text has been obtained from the patient in line with COPE guidance.
[fig] Figure 1: Contrast agent extravasation following perforation of a pulmonary artery side branch (A) with subsequent balloon occlusion (B). [/fig]
[fig] Figure 2: Selective visualization of the perforation cavity via the inserted microcatheter (A). Complete embolization of the perforation cavity after implantation of four coils (B). Iatrogenic perforation of a pulmonary artery side branch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 mmHg, perhaps due to the increase in diuretic doses compared with the day of hospital admission. [/fig]
[fig] Figure 3: Computed tomography scan showing coils in the left pulmonary artery. Pulmonary artery is marked in blue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [/fig]
|
Between a hygiene rock and a hygienic hard place
# Introduction
Coronavirus disease 2019 (COVID- [bib_ref] The helminth parasite Heligmosomoides polygyrus attenuates EAE in an IL-4Ra-dependent manner, White [/bib_ref] is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a single stranded RNA virus in the Coronaviridae family. In the first year following the World Health Organization's declaration of the disease a Public Health Emergency of International Concern on 30 January 2020, the disease claimed >2 million lives worldwide. Much of the reported mortality was centered in particular high-income countries. For example, based on data collected by the Johns Hopkins Coronavirus Resource Center as of January 2021, the USA and the UK together accounted for more than one quarter of all fatalities during the first year, despite accounting for only about 5% of the world's total population.
During the first months of the COVID-19 pandemic, data began to emerge connecting adverse and often deadly reactions to the SARS-CoV-2 virus with autoimmune-like reactions (reviewed by Halpert and Shoenfeld. Based on that information, and based on the known impact of the loss of symbiotic intestinal worms (helminths) associated with hygiene, one of us made a prediction in May of 2020, posted on social media, that infection with the SARS-CoV-2 virus may not be as deadly in parts of the world without widespread use of toilets and water treatment facilities (www.facebook.com/WilliamParkerLab, last accessed . This prediction is now supported by several comparative studies examining the impact of COVID-19 in different parts of the world [bib_ref] The enigma of low COVID-19 fatality rate in India, Samaddar [/bib_ref] [bib_ref] Differential mortality in COVID-19 patients from India and western countries, Jain [/bib_ref] [bib_ref] CoV-2 infection in India bucks the trend: trained innate immunity?, Chinnaswamy [/bib_ref] [bib_ref] High SARS-CoV-2 seroprevalence in health care workers but relatively low numbers of..., Chibwana [/bib_ref].
In this review, we discuss hygiene and its multiple impacts on human health. Particular regard is given to the impact of hygiene on the COVID-19 pandemic. On the one hand, hygiene helps alleviate very high burdens of infectious disease that result from increased population densities that, in turn, occurred as a result of the development of agriculture and urban centers. For example, fortunately, hygiene measures help reduce the morbidity and mortality resulting from the COVID-19 pandemic. On the other hand, hygiene causes loss of some species from the ecosystem of the human body, which in turn leads to susceptibility to a range of chronic inflammatory disease. Unfortunately, this susceptibility to chronic inflammatory disease is apparently associated with adverse reactions to a large number of viral pathogens, probably including SARS-CoV-2.
In the first section of this review, a brief history of our understanding of the impact of hygiene on human health is provided. Hygiene will be divided into two categories based on how hygiene is implemented in society and based on the impact of that hygiene. Next, the impact of the two types of hygiene on the COVID-19 pandemic will be discussed. Finally, we will discuss potential solutions to the problem as well as hurdles that currently impede implementation of those solutions. Text boxes summarizing this information are provided to function as stand-alone, public-domain handouts, both for informing the general public about the importance of hygiene and to provide useful information for biomedical research scientists and policy makers involved with our response to the COVID-19 pandemic.
## The dark side of hygiene
Although it is widely known that hygiene must be employed to slow or in some cases prevent pandemics of infectious disease, not all effects of hygiene are positive. The term hygiene hypothesis was first coined by David Barker in 1988, who observed that improved sanitation conditions are associated with appendicitis [bib_ref] Appendicitis epidemic following introduction of piped water to Anglesey, Barker [/bib_ref]. A year later, David Strachan used the term again, suggesting that increased exposures to a wide range of pathogens might decrease the likelihood of allergic disease [bib_ref] Hay fever, hygiene, and household size, Strachan [/bib_ref]. The hygiene hypothesis was derided at the time because no plausible mechanism that might explain the hypothesis had been discovered [bib_ref] The "hygiene hypothesis" for allergic disease is a misnomer, Parker [/bib_ref]. However, both Barker and Strachan's work proved to be seminal and is now understood in terms of improved immune regulation as a result of exposure to particular types of organisms such as intestinal worms, called helminths [bib_ref] Regulatory T cells induced by parasites and the modulation of allergic responses, Wilson [/bib_ref] [bib_ref] Infections and allergy -helminths, hygiene and host immune regulation, Maizels [/bib_ref] [bib_ref] Th2 responses without atopy: immunoregulation in chronic helminth infections and reduced allergic..., Yazdanbakhsh [/bib_ref].
Unknown to Barker and Strachan when they published their studies, Brian Greenwood, more than a decade earlier, had hypothesized that infections with helminths and protists could explain the lack of autoimmune disease he observed in Ibadan, Nigeria [bib_ref] Autoimmune disease and parasitic infections in Nigerians, Greenwood [/bib_ref] [bib_ref] Voller A; Can parasitic infection suppress autoimmune disease?, Greenwood [/bib_ref]. Greenwood quickly went on to demonstrate that a mild infection with protists, which produced no apparent long-term adverse effects, completely rescued laboratory mice that would otherwise die from a lupus-like autoimmune condition [bib_ref] Voller A; Suppression of autoimmune disease in NZB and (NZB x NZW)..., Greenwood Bm [/bib_ref]. At the same time, again via exposure to protists, Greenwood was able to block autoimmune disease in a laboratory rat model of rheumatoid arthritis [bib_ref] Suppression of adjuvant arthritis by infection with a strain of the rodent..., Greenwood [/bib_ref]. Soon afterward, Peter John Preston observed that exposure to helminths effectively alleviates symptoms of seasonal allergies [bib_ref] The biology of the atopic response, Preston [/bib_ref]. Preston's observation was confirmed by a self-infection experiment conducted by Jon Turton, described in 1976 [bib_ref] IgE, parasites, and allergy, Turton [/bib_ref]. More recently, studies in animal models by Maizels et al. [bib_ref] The helminth parasite Heligmosomoides polygyrus attenuates EAE in an IL-4Ra-dependent manner, White [/bib_ref] [bib_ref] Immunomodulation of experimental autoimmune encephalomyelitis by helminth ova immunization, Sewell [/bib_ref] as well as prospective clinical studies by Correale et al. [bib_ref] Helminth infections associated with multiple sclerosis induce regulatory B cells, Correale [/bib_ref] [bib_ref] Association between parasite infection and immune responses in multiple sclerosis, Correale [/bib_ref] [bib_ref] The impact of parasite infections on the course of multiple sclerosis, Correale [/bib_ref] have extended the apparently beneficial role of exposure to helminths to include prevention of autoimmune conditions [bib_ref] van Ree R; Allergy, parasites, and the hygiene hypothesis, Yazdanbakhsh [/bib_ref] [bib_ref] Helminth immunomodulation in autoimmune disease, Smallwood [/bib_ref]. Further, emerging evidence indicates that exposure to helminths also helps prevent or alleviate a range of inflammationassociated neuropsychiatric disorders, including chronic fatigue syndrome, anxiety disorders and major depressive disorders [bib_ref] Got worms? Perinatal exposure to helminths prevents persistent immune sensitization and cognitive..., Williamson [/bib_ref] [bib_ref] Overcoming evolutionary mismatch by self-treatment with helminths: current practices and experience, Cheng [/bib_ref] [bib_ref] Practices and outcomes of selftreatment with helminths based on physicians' observations, Liu [/bib_ref] [bib_ref] Production and use of Hymenolepis diminuta cysticercoids as anti-inflammatory therapeutics, Smyth [/bib_ref].
At the present time, the state of our understanding can be described as a Biota Alteration Theory, whereby a hygiene-induced lack of biodiversity in post-industrial society, including the essentially complete loss of complex eukaryotic symbionts such as helminths and protists, poses a major evolutionary mismatch [bib_ref] A prescription for clinical immunology: the pills are available and ready for..., Parker [/bib_ref] [bib_ref] Evolutionary biology and anthropology suggest biome reconstitution as a necessary approach toward..., Parker [/bib_ref]. This evolutionary mismatch, or point of incompatibility between our biology and modern society, creates generalized immune dysfunction and predisposes humans to a wide range of inflammation-associated maladies [bib_ref] A prescription for clinical immunology: the pills are available and ready for..., Parker [/bib_ref] [bib_ref] Evolutionary biology and anthropology suggest biome reconstitution as a necessary approach toward..., Parker [/bib_ref] [bib_ref] Review series on helminths, immune modulation and the hygiene hypothesis: the broader..., Rook [/bib_ref] [bib_ref] Microbial 'old friends', immunoregulation and socioeconomic status, Rook [/bib_ref]. This adverse result of the loss of complex eukaryotic symbionts can be readily understood in light of the hundreds of millions of years of vertebrate evolution in the presence of helminths [bib_ref] Reconstitution of the human biome as the most reasonable solution for epidemics..., Bilbo [/bib_ref]. In short, the vertebrate immune system has evolved in the presence of helminths and protists, and as a result is now dependent on exposure to these organisms for effective development and function [bib_ref] Review series on helminths, immune modulation and the hygiene hypothesis: immunity against..., Jackson [/bib_ref].
Although hygiene causes one evolutionary mismatch by eliminating symbiotic organisms that are necessary for efficient immune function, it is widely appreciated that hygiene alleviates the consequences of another evolutionary mismatch. In particular, hygiene, in combination with vaccine programs, mitigates the high burdens of infectious disease that result from crowded, urban environments. These crowded environments and their associated burden of infection constitute an evolutionary mismatch that is very distinct from but yet tied closely to the evolutionary mismatch of biota alteration.
## Personal hygiene versus systems hygiene
Importantly, the seminal work by Greenwood, Preston, Turton, Barker, Strachan, Maizels, Correale and others that led to our current understanding was associated with a type of hygiene that is no longer thought of as hygiene, per se, in most parts of post-industrial, high-income countries. Hygiene today in highincome countries is not defined by whether we use a toilet, whether we have refrigeration and plastic storage containers, whether our food is harvested and prepared using modern technology, or whether we have water from municipal sources. These factors create a type of hygiene, which we will label systems hygiene. This type of hygiene is indeed necessary to prevent the spread of infectious disease and also causes biota alteration and the ensuing propensity for chronic, noninfectious and inflammation-associated diseases. But this is not that type of hygiene that those of us in industrial societies tend to think about when the term hygiene is mentioned.
When people think of hygiene in high-income countries, they think of personal hygiene, the type of hygiene that is not at the root of biota alteration. Personal hygiene is well recognized by washing hands, brushing teeth, taking showers and appropriate social distancing when needed. This type of hygiene, as with systems hygiene, is necessary for the prevention and control of pandemics of infectious disease. However, unlike systems hygiene that leads to biota depletion and subsequent chronic inflammatory disease, personal hygiene actually decreases acute inflammatory stressors such as mold, insect-derived and mite-derived allergens and acute viral infections, all of which can trigger chronic inflammatory disease in an immune system destabilized by biota alteration. Thus, personal hygiene, unlike systems hygiene, can reduce the burden of chronic, non-infectious, inflammation-associated disease. A summary of the As shown in [fig_ref] Figure 1: Two types of hygiene with sometimes overlapping and sometimes opposing effects [/fig_ref] and summarized in Box 1, it is not tenable to compensate for biota alteration by eliminating either systems hygiene or personal hygiene. Elimination of either type of hygiene would result in catastrophic morbidity and mortality from infectious disease. Further, relaxing personal hygiene would likely only increase exposure to triggers of chronic inflammatory disease without alleviating the biota alteration induced by systems hygiene.
## The impact of biota alteration on disease and the covid-19 pandemic
The effects of biota alteration began to emerge in the late 19 th century and continue to expand to this day. Some of the first effects were documented toward the latter part of the 19 th century and include seasonal allergiesand appendicitis [bib_ref] The functional nature of the caecum and appendix, Keith [/bib_ref]. Although the pathological effects of biota alteration can be diverse, they share several factors in common. The diseases tend to be chronic, difficult to cure using pharmaceutical approaches and are always associated with inflammation. Further, although the evolutionary mismatch of biota alteration predisposes individuals to chronic inflammatory disease, the induction of disease usually depends on a wide range of other factors, including genetics and exposure to triggers that can induce disease. For example, while biota alteration is at the root of seasonal allergies [bib_ref] The biology of the atopic response, Preston [/bib_ref] , such allergies are associated with genetic predisposition as well as exposure to allergens such as ragweed pollen and dust mite-derived proteins.
One of the most pronounced effects of widespread biota alteration on a society is the dramatic increase in the prevalence of autoimmune disease. Autoimmune diseases, now roughly 100 in number, affect >8% of the total US population. Unfortunately, as shown in [fig_ref] Figure 2: The development of a wide range of autoimmune diseases stemming from all... [/fig_ref] , a common trigger for the induction of autoimmune disease is viral infection. Indeed, all major Baltimore classes of viruses, including major classes of RNA and DNA viruses, are thought to induce autoimmune disease in humans [fig_ref] Figure 2: The development of a wide range of autoimmune diseases stemming from all... [/fig_ref]. Thus, while exposure to SARS-CoV-2 may induce autoimmune disease [bib_ref] Guillain-Barré syndrome: the first documented COVID-19-triggered autoimmune neurologic disease: more to come..., Dalakas [/bib_ref] [bib_ref] Covid-19 and autoimmunity, Ehrenfeld [/bib_ref] the virus is not unusual in this regard.
Another hallmark of high-income countries is the high prevalence of inflammation-associated neuropsychiatric disorders. For example, >1% of US women are affected by chronic fatigue syndrome [bib_ref] Systematic review and meta-analysis of the prevalence of chronic fatigue syndrome/myalgic encephalomyelitis..., Lim [/bib_ref] [bib_ref] Inflammation correlates with symptoms in chronic fatigue syndrome, Komaroff [/bib_ref] and as many as 20% of US adults may be affected by major depressive disorder [bib_ref] Altered expression of genes involved in inflammation and apoptosis in frontal cortex..., Shelton [/bib_ref] [bib_ref] Depression: an inflammatory illness?, Krishnadas [/bib_ref] [bib_ref] Epidemiology of adult DSM-5 major depressive disorder and its specifiers in the..., Hasin [/bib_ref]. Neuropsychiatric Box 1. Clearing the cloud of confusion surrounding hygiene: two types of hygiene with different effects on COVID-19 and on our immune system 1. The term hygiene can refer to either one of two very different ways of avoiding infection. The types of hygiene, labeled here as personal hygiene and systems hygiene, have very different effects on humans. 2. Personal hygiene, including hand washing and social distancing, helps prevent transmission of many infectious diseases, including COVID-19 and the flu. Diseases prevented by personal hygiene are often dangerous and detrimental to immune function. 3. Improved personal hygiene can reduce exposure to a range of factors that can trigger chronic inflammatory conditions. Such inflammation-inducing factors include mold, insect-derived allergens and acute infections from a wide range of viruses that includes SARS-CoV-2. 4. Systems hygiene is the implementation of modern sanitation (water treatment plants, sewage systems, indoor plumbing) and food processing and storage technology. Systems hygiene, like personal hygiene, also helps prevent transmission of many infectious diseases that are unhelpful for immune function and detrimental to health. For example, systems hygiene prevents pandemics of typhoid, cholera and amoebic dysentery. 5. Systems hygiene, however, causes a virtually complete loss of some types of organisms that have lived in the bodies of our ancestors for hundreds of millions of years. Intestinal worms, called helminths, are one of the key species that have been all but driven extinct by systems hygiene in modern society. Even though many helminth species are harmful parasites, others cause little or no disease. Importantly, scientific evidence demonstrates that exposure to helminths appears to be necessary for healthy immune function and that their absence leaves us susceptible to pandemics of chronic, non-infectious inflammatory diseases. Such diseases include allergy, autoimmunity and probably neuropsychiatric issues such as major depression, anxiety disorders and chronic fatigue syndrome. 6. Reductions in personal hygiene cannot reverse the detrimental loss of helminths caused by systems hygiene. Rather, reductions in personal hygiene would only cause more infectious diseases such as COVID-19 and would result in more exposure to inflammation-inducing factors, all of which would be harmful. 7. We cannot live without systems hygiene. Loss of systems hygiene in modern society would, in theory, reverse the detrimental loss of helminths caused by that type of hygiene. However, the resulting waves of deadly, infectious diseases would be catastrophic.
disorders, like autoimmune diseases, are probably associated with biota alteration [bib_ref] Got worms? Perinatal exposure to helminths prevents persistent immune sensitization and cognitive..., Williamson [/bib_ref] [bib_ref] Overcoming evolutionary mismatch by self-treatment with helminths: current practices and experience, Cheng [/bib_ref] [bib_ref] Practices and outcomes of selftreatment with helminths based on physicians' observations, Liu [/bib_ref] [bib_ref] Production and use of Hymenolepis diminuta cysticercoids as anti-inflammatory therapeutics, Smyth [/bib_ref] [bib_ref] Rook GW; Inflammation, sanitation, and consternation: loss of contact with coevolved, tolerogenic..., Raison [/bib_ref] [bib_ref] Microbiota, immunoregulatory old friends and psychiatric disorders, Rook [/bib_ref] [bib_ref] Intestinal worms eating neuropsychiatric disorders? Apparently so, Kou [/bib_ref]. Although the idea that viral infection can trigger neuropsychiatric disorders is difficult to assess, some evidence exists [bib_ref] Viral infection and neurological disorderspotential role of extracellular nucleotides in neuroinflammation, Li [/bib_ref] [bib_ref] Role of infection in neurologic and psychiatric diseases, Bortolato [/bib_ref]. Again, SARS-CoV-2 is possibly not unique in this regard, triggering a variety of neuropsychiatric issues including delirium, behavioral changes and encephalopathy [bib_ref] Neuropsychiatric manifestations of COVID-19 and possible pathogenic mechanisms: insights from other coronaviruses, Banerjee [/bib_ref]. However, given limited data at the present time, it remains unknown whether SARS-CoV-2 can trigger long term neuropsychiatric issues. Emerging evidence indicates that much of the morbidity/ mortality associated with SARS-CoV-2 is in fact due to an overly aggressive immune response and ensuing cytokine storm [bib_ref] Biochemical indicators of coronavirus disease 2019 exacerbation and the clinical implications, An [/bib_ref] [bib_ref] 19: review on latest available drugs and therapies against SARS-CoV-2. Coagulation and..., Magro [/bib_ref] [bib_ref] The cytokine storm of COVID-19: a spotlight on prevention and protection, Pearce [/bib_ref] [bib_ref] Cytokine storm induced by SARS-CoV-2, Song [/bib_ref] that may be partly autoimmune in nature [bib_ref] The cytokine storm of COVID-19: a spotlight on prevention and protection, Pearce [/bib_ref]. In general, in the field of medicine, the presence of helminths is thought to produce an attenuated immune response due to secretion of immunoregulatory molecules [bib_ref] Molecules of parasites as immunomodulatory drugs, Imai [/bib_ref] , creation of regulatory networks [bib_ref] Regulatory T cells induced by parasites and the modulation of allergic responses, Wilson [/bib_ref] and changes in mucosal surface permeability [bib_ref] Helminthic therapy: improving mucosal barrier function, Wolff [/bib_ref]. However, given that vertebrate immune function evolved in the presence of symbiotic helminths [bib_ref] Reconstitution of the human biome as the most reasonable solution for epidemics..., Bilbo [/bib_ref] [bib_ref] Review series on helminths, immune modulation and the hygiene hypothesis: immunity against..., Jackson [/bib_ref] , the field of medicine needs a change in perspective and may benefit greatly by considering the immune response in the presence of helminths to be 'normal' and function in the absence of helminths to be hyperresponsive [bib_ref] Review series on helminths, immune modulation and the hygiene hypothesis: the broader..., Rook [/bib_ref] [bib_ref] Regulation of allergy and autoimmunity in helminth infection, Wilson [/bib_ref]. With this view in mind, it seems intuitive that biota alteration may be, at least in part, responsible for part of the morbidity and mortality associated with SARS-CoV-2 in high-income countries. [bib_ref] A comprehensive and quantitative exploration of thousands of viral genomes, Mahmoudabadi [/bib_ref]. The development of RA after Chikungunya virus infection is described by Tanay [bib_ref] Chikungunya virus and autoimmunity, Tanay [/bib_ref] , T1D after the mumps virus by Ramondetti et al. [bib_ref] Type 1 diabetes and measles, mumps and rubella childhood infections within the..., Ramondetti [/bib_ref] , Sjögren's syndrome after HTLV-I by Quaresma et al. [bib_ref] HTLV-1, Immune response and autoimmunity, Quaresma [/bib_ref] and autoimmune diseases after HBV by Maya et al. [bib_ref] Hepatitis B virus (HBV) and autoimmune disease, Maya [/bib_ref]. The development of GBS after SARS-CoV-2 infection is described by and MFS and KD/KDSS after SARS-CoV-2 by Ehrenfeld et al. [bib_ref] Covid-19 and autoimmunity, Ehrenfeld [/bib_ref]. The development of all other autoimmune conditions after corresponding viral infections was recently summarized by Smatti et al. [bib_ref] Viruses and autoimmunity: a review on the potential interaction and molecular mechanisms, Smatti [/bib_ref]
## Systems hygiene and covid-19
The effects of hygiene on the COVID-19 pandemic are potentially complicated (Box 2). However, based on the effects of biota alteration on immune function and based on emerging data regarding the immune response to COVID-19 in those countries, one of us predicted in May of 2020 that COVID-19 would likely be less impactful in low-income countries than in high-income countries (Facebook post from Dr. William Parker's Lab, last accessed December 17, 2020). This prediction was made despite the fact that social distancing and cessation of many business practices may be less feasible in lowincome countries than in other countries. Emerging evidence now indicates that people living in areas with endemic exposure to helminths are in fact less impacted by infection with SARS-CoV-2 than are individuals living in high-income countries [bib_ref] The enigma of low COVID-19 fatality rate in India, Samaddar [/bib_ref] [bib_ref] Differential mortality in COVID-19 patients from India and western countries, Jain [/bib_ref] [bib_ref] CoV-2 infection in India bucks the trend: trained innate immunity?, Chinnaswamy [/bib_ref] [bib_ref] High SARS-CoV-2 seroprevalence in health care workers but relatively low numbers of..., Chibwana [/bib_ref]. Numerous post-hoc explanations for this observation that surprised most infectious disease experts have been offered [bib_ref] The enigma of low COVID-19 fatality rate in India, Samaddar [/bib_ref] , including the presence of relatively fewer elderly and chronically ill individuals in low-income countries. While demographics apparently do account for some of the less severe effects of SARS-CoV-2 in low-income countries, the decreased levels of systems hygiene and autoimmune disease also appear to be associated with the reduced impact of SARS-CoV-2 in countries such as the Democratic Republic of Congo, the Republic of the Philippines and the Republic of Haiti [bib_ref] Helminth coinfection and COVID-19: an alternate hypothesis, Hays [/bib_ref] [bib_ref] COVID-19 lethality in Sub-Saharan Africa and helminth immune modulation, Fonte [/bib_ref].
Given the complexity of the pandemic, experts working in the field realize that the consensus at the present time is tenuous. In addition, current studies use measures of income, waterborne diseases, prevalence of autoimmunity or other factors that do not necessarily correlate exactly with the presence of helminths and may not be constant across a given country, potentially clouding the data. Nevertheless, if the initial conclusions drawn from early studies prove to be correct, it strongly supports the view that restoring the biota in high-income countries will not only decrease the burden of autoimmune and allergic disease but might also decrease the damage potentially caused by future pandemics of infectious disease.
## Restoring the biota: where we are now
More than 50 years ago, Brian Greenwood deduced that high parasite burdens were probably responsible for very low levels of autoimmune disease in some communities [bib_ref] Autoimmune disease and parasitic infections in Nigerians, Greenwood [/bib_ref] and subsequently demonstrated that exposure to a protist could in fact rescue laboratory mice from a lethal, lupus-like syndrome [bib_ref] Voller A; Suppression of autoimmune disease in NZB and (NZB x NZW)..., Greenwood Bm [/bib_ref]. Importantly, Greenwood's 'protist therapy' produced healthy mice with no apparent lingering side effects from the life-saving therapy, thus demonstrating that controlled exposure to a symbiont could in fact be highly beneficial. Although the laboratory mouse model used by Greenwood is still popular today [bib_ref] Murine models of systemic lupus erythematosus, Perry [/bib_ref] , his cure did not garner interest from the scientific community. Similarly, studies in the 1970s by Preston [bib_ref] The biology of the atopic response, Preston [/bib_ref] and Turton [bib_ref] IgE, parasites, and allergy, Turton [/bib_ref] showing that exposure to helminths eliminated seasonal allergies was never pursued by the medical community. It was not until about 2003, when Weinstock et al. [bib_ref] Helminths and the modulation of mucosal inflammation, Elliott [/bib_ref] [bib_ref] Trichuris suis therapy in Crohn's disease, Summers [/bib_ref] began controlled exposures of patients to the porcine whipworm in an effort to treat inflammatory bowel disease, that interest in developing a clinical therapy using symbionts picked up.
Studies evaluating the effects of protists and helminths on multiple sclerosis [bib_ref] The impact of parasite infections on the course of multiple sclerosis, Correale [/bib_ref] suggest that a wide range of complex eukaryotic symbionts may achieve a similar therapeutic effect. However, since the reproduction of helminths is much more easily controlled than is the reproduction of protists, helminths rather than protists are preferred for development as a therapeutic. Systematic sociomedical studies evaluating the effects of helminth therapy on individuals engaging in self-treatment with helminths confirms studies in laboratory animals and indicates that the therapy is indeed very effective and safe for a range of common allergic, autoimmune and neuropsychiatric conditions [bib_ref] Overcoming evolutionary mismatch by self-treatment with helminths: current practices and experience, Cheng [/bib_ref] [bib_ref] Practices and outcomes of selftreatment with helminths based on physicians' observations, Liu [/bib_ref] [bib_ref] Production and use of Hymenolepis diminuta cysticercoids as anti-inflammatory therapeutics, Smyth [/bib_ref].
Unfortunately, despite some ongoing work in the field using the human hookworm as a therapeutic agent, research is Box 2. The complex relationship between hygiene and COVID-19
1. Personal hygiene (social distancing, handwashing and wearing a mask that covers the mouth and nose) is the best approach to avoiding SARS-CoV-2 and the devastating effects of the COVID-19 pandemic. 2. Systems hygiene (sewer systems, water treatment systems and food processing and storage facilities) reduces or even eliminates exposure to intestinal worms, called helminths. However, emerging evidence suggests that human populations with helminths are less prone to have deadly or severe adverse reactions to SARS-CoV-2. 3. Many viruses, including SARS-CoV-2, can trigger autoimmune reactions and/or diseases. Nevertheless, it is the loss of helminths in the body that predisposes the immune system to develop autoimmune diseases and allergic disorders. 4. High-income countries are stuck between a hygiene rock and a hygienic hard place: they need controlled exposures to helminths to ensure effective immune function, but at the same time they need effective systems hygiene to avoid environmental exposures to deadly organisms such as typhoid or cholera.
moving slowly without any unified effort, perhaps due to the same lack of attention that Greenwood's work [bib_ref] Evolution of the hygiene hypothesis into biota alteration theory: what are the..., Villeneuve [/bib_ref] faced more than half a century ago. Key problems with garnering interest include policy and regulatory issues that greatly diminish commercial incentives for developing naturally occurring symbionts as a therapy [bib_ref] Policy and regulations in light of the human body as a 'superorganism'..., Bono-Lunn [/bib_ref] [bib_ref] Prerequisites for the pharmaceutical industry to develop and commercialise helminths and helminth-derived..., Tilp [/bib_ref]. In addition, the false bias that all helminths are harmful parasites tends to dissuade many from ever considering the idea [bib_ref] Not infection with parasitic worms, but rather colonization with therapeutic helminths, Parker [/bib_ref]. Bias against helminths is understandable, but evidence suggests that considering all helminths to have the same effects on the human body would be the same as lumping salmonella and lactobacilli into a single group simply because they are both bacteria. For example, while some helminths can cause cancer [bib_ref] Why does infection with some helminths cause cancer?, Brindley [/bib_ref] , others have been shown to prevent cancer in animal models [bib_ref] Extraintestinal helminth infection reduces the development of colitis-associated tumorigenesis, León-Cabrera [/bib_ref].
Another factor that has diminished interest in using helminths as a therapeutic agent is the unsubstantiated idea that helminth derived products rather than intact helminths might be effective therapeutics. This idea has garnered much interest because it promises financial rewards, but it seems highly unlikely that a single molecule, whether delivered via injection or by pill, could recapitulate the complex biological relationships between helminth and host that effectively modulate immune function [bib_ref] A prescription for clinical immunology: the pills are available and ready for..., Parker [/bib_ref]. Further, lack of interest in the field is likely encouraged by lackluster results of clinical studies that, although time consuming and costly, were probably not designed appropriately to evaluate the use of helminths as a therapeutic agent. For example, two primary concerns with prior studies are that (i) they do not take into account large variations in the amount of exposure to helminths necessary for beneficial effects in humans and (ii) they do not take into account potential nuances in the production and storage of living organisms that might affect their clinical efficacy [bib_ref] Overcoming evolutionary mismatch by self-treatment with helminths: current practices and experience, Cheng [/bib_ref]. A further complicating problem is that the selection of helminth species for use in clinical trials thus far has been haphazard, dictated by convenience rather than by a systematic search for a benign helminth that effectively regulates immune function [bib_ref] Helminth therapy -from the parasite perspective, Sobotková [/bib_ref].
Perhaps the best indicator of impediments facing the use of naturally occurring organisms to effectively treat disease is the history of the fecal microbiota transplant. First established prior to 1960 as an effective means of treating a common but deadly condition known as recurrent Clostridium difficile colitis [bib_ref] Fecal enema as an adjunct in the treatment of pseudomembranous enterocolitis, Eiseman [/bib_ref] , the procedure did not become popular until 40 years later, only after tens of thousands of patients with recurrent C. difficile colitis died without the life-saving transplant [bib_ref] Policy and regulations in light of the human body as a 'superorganism'..., Bono-Lunn [/bib_ref]. Sadly, the transplant procedure is still not standard of care today, with unknown numbers of patients still dying from recurrent C. difficile colitis who might have been saved by fecal material from a healthy donor. This scenario highlights the problems currently facing helminth therapy and biota reconstitution in general. Without pharmaceutical-based incentives to drive medical practice, the modern medical enterprise falters. In essence, the concept of evolutionary mismatch applies: The system that evolved with a foundation of pharmaceutical involvement does not perform well when that underpinning is removed.
## Conclusions: problems and solutions
Independent lines of evidence, including epidemiologic studies, studies using animal models and clinical observations, point to the idea that we need exposure to helminths in order to avoid biota alteration and immune hypersensitivity. Further, prior Box 3. The path forward for medicine: how we can avoid infectious disease and still obtain enough environmental exposures to support our immune system? 1. The detrimental effects of the loss of intestinal worms, called helminths, caused by factors such as toilets and water treatment facilities can be readily reversed by domestication of select helminth species and artificial enrichment of the human body with those organisms. Considerable evidence supports the view that such an effort would greatly reduce the burden of allergy, autoimmunity and probably neuropsychiatric disorders currently experienced in high-income countries. It also seems very likely that these efforts would decrease the likelihood of having adverse reactions to infections with a wide range of viruses, including SARS-CoV-2. 2. Work in the field of helminth therapy indicates that current trials based on pharmaceutical models fail to take into account critical issues, including individual-to-individual variation in the effective dose, risk/benefit ratios when selecting helminth species and the importance of specific husbandry conditions when cultivating and preserving helminths. Trials need to be conducted with appropriate methods of production of helminths, dosing regimens designed for helminth therapy and rational selection of helminth species. 3. Although benign helminths are currently available for human testing, interest in conducting clinical trials is hampered by high costs and intellectual property issues. Given that therapies based on naturally occurring organisms cannot be patented, financial incentives for moving forward are lacking. 4. In general, governments and research organizations need to focus on disease prevention by dealing with evolutionary mismatches rather than treatment of disease solely using pharmaceutical approaches. This principle applies to many facets of modern medicine, including how we deal with the adverse effects of the loss of helminths on our immune system. studies indicate that our response to acute viral infections may be exacerbated by biota alteration, and emerging evidence suggests that this paradigm also applies to the immune response against SARS-CoV-2. Although vaccines against SARS-CoV-2 are forthcoming, the virus may become endemic and seasonal [bib_ref] Will SARS-CoV-2 become endemic?, Shaman [/bib_ref] , providing further impetus for efforts aimed at reversing the effects of biota alteration in high-income countries. In addition, effective immune function is undoubtedly important when considering current pandemics of autoimmune disease, allergy, neuropsychiatric disorders and susceptibility to as yet unknown pandemics induced by viral infections that might occur in the future. At the same time, eliminating systems hygiene to alleviate the effects of biota alteration would induce unacceptable pandemics of infectious disease. Eliminating personal hygiene would exacerbate current problems with infectious disease, increase exposure to agents that trigger chronic disease and utterly fail to relieve the adverse effects of biota alteration.
Fortunately, a course of action is available (Box 3). Clinical trials need to be conducted to determine how to alleviate the adverse effects of biota alteration. Helminths, like bacteria, are not all pathogens or parasites [bib_ref] Evolutionary biology and anthropology suggest biome reconstitution as a necessary approach toward..., Parker [/bib_ref] [bib_ref] Helminth therapy -from the parasite perspective, Sobotková [/bib_ref] , despite our prejudice against them. Indeed, benign helminths are already available which can be evaluated for their potential to reverse the effects of biota alteration [bib_ref] A prescription for clinical immunology: the pills are available and ready for..., Parker [/bib_ref] [bib_ref] Production and use of Hymenolepis diminuta cysticercoids as anti-inflammatory therapeutics, Smyth [/bib_ref] [bib_ref] Helminth therapy -from the parasite perspective, Sobotková [/bib_ref] [bib_ref] Helminth infections decrease host susceptibility to immune-mediated diseases, Weinstock [/bib_ref]. Unfortunately, clinical trials with 'helminth therapy' to date have been problematic, not taking into account nuances with the cultivation of helminths, selection of helminths with the best benefit-to-risk ratios and the wide range of individual-to-individual variation in the effective treatment regimens [bib_ref] Overcoming evolutionary mismatch by self-treatment with helminths: current practices and experience, Cheng [/bib_ref]. Action is urgently needed given the importance of immune function, not only for immunological tolerance to our own bodies and harmless environmental antigens, but also for effective, self-preserving responses against current and future pathogens. Unfortunately, logical and reasonable courses of action, summarized in Box 3, have largely been hampered by current regulatory pathways and financial incentives that encourage drug development but impede more straight-forward reversal of evolutionary mismatches such as biota alteration [bib_ref] Evolution of the hygiene hypothesis into biota alteration theory: what are the..., Villeneuve [/bib_ref] [bib_ref] Policy and regulations in light of the human body as a 'superorganism'..., Bono-Lunn [/bib_ref] [bib_ref] Prerequisites for the pharmaceutical industry to develop and commercialise helminths and helminth-derived..., Tilp [/bib_ref]. With this in mind, questions regarding the future of immune function in high-income countries (Box 4) point toward much needed changes in policy and regulation to incentivize work on restoration of a healthy biota.
acknowledgements This article was stimulated in part by discussions with A.E. van der Meulen (biochemist) on interpretations of hygiene notions for SARS-CoV-2 and other pathogens in the COVID-19 era among the general public, including Box 4. Frequently asked questions regarding COVID-19 and hygiene 1. Will one or two years of social distancing for COVID-19 'crash our immune system'? No. The loss of particular organisms from within our bodies due to widespread use of sewage treatment facilities and water purification plants does indeed damage immune function and could lead to chronic inflammatory diseases and probably adverse reactions to viruses such as SARS-CoV-2. However, reducing personal hygiene would have no effect on this problem. 2. Is it possible to restore specific lost species in our body and simultaneously maintain hygiene to avoid pandemics of infectious disease? The answer is absolutely yes-in theory. There is no evidence to suggest that we need to be exposed to disease causing organisms to have the appropriate array of organisms in our body for healthy immune function. We can domesticate the organisms we need and introduce them artificially. Long-standing evidence indicates that exposure to selected intestinal worms will be effective at reducing disease without causing health problems, if that solution can be implemented. 3. Why haven't we already corrected the problem by restoring organisms that have been lost? The answer to this question is multifaceted. First, even though the adverse effects of missing particular organisms have been mounting for over a century, we only recently understood the problem well enough to do something about it. Second, unfortunately, organisms such as intestinal worms, called helminths, that could be used to restore the ecosystem of the body, are classified as drugs under current regulations. Classification as a drug presents a number of problems, including a drug development process unsuited for naturally occurring organisms such as helminths. Third, the limited trials that have taken place thus far using helminths to restore the biodiversity of the body did not take into account several important factors that need to be considered when designing trials using helminths as therapeutic agents and thus did not move the field forward. 4. How long will it take to restore the needed species in our bodies? The answer is that, since candidate helminths have already been identified and are currently present in the environment without causing disease, products for consumer use could be available within a year or two of the initiation of well-designed trials. However, we do not yet know when policy makers and health officials will address the issues that are currently impeding research in this area. Thus, the technology is available to solve the problem quickly, but it is difficult to predict when necessary policy and funding changes will be made. 5. What about the gut microbiota, the bacteria? Can we use bacteria to help restore healthy immune function to our bodies?
The answer is unknown. Our diets are the major drivers of the microbiota community composition, and we don't know if it's possible to fix our microbiota without fixing our diets. Further, we do not know how important alterations in the microbiota are in terms of a causative agent of chronic inflammatory disease.
individuals with advanced biomedical training. We apologize to colleagues whose work was not cited due to space constraints. Co-author W.P. is grateful to the non-profit foundation Immunity's Forge and to Hiroko
[fig] Figure 1: Two types of hygiene with sometimes overlapping and sometimes opposing effects. Triggers of chronic inflammatory disorders that are reduced by personal hygiene include viral infections, dust mite derived allergens and mold 122 | Parker et al. Evolution, Medicine, and Public Health two types of hygiene and their corresponding effects are shown in Fig. 1 and in Box 1. [/fig]
[fig] Figure 2: The development of a wide range of autoimmune diseases stemming from all seven Baltimore classifications of viruses. Baltimore classifications for viruses are from Mahmoudabadi and Phillips [/fig]
[bib_ref] The helminth parasite Heligmosomoides polygyrus attenuates EAE in an IL-4Ra-dependent manner, White [/bib_ref] |
Interventions for improving management of chronic non-communicable diseases in Dikgale, a rural area in Limpopo Province, South Africa
Background: Chronic disease management (CDM) is an approach to health care that keeps people as healthy as possible through the prevention, early detection and management of chronic diseases. The aim of this study was to develop interventions to improve management of chronic diseases in the form of an integrated, evidence-based chronic disease management model in Dikgale, a rural area of Limpopo Province in South Africa. Methods: A multifaceted intervention, called 'quality circles' (QCs) was developed to improve the quality and the management of chronic diseases in the Dikgale Health and Demographic Surveillance System (HDSS). These QCs used the findings from previous studies which formed part of the larger project in the study area, namely, the quantitative study using STEPwise survey and qualitative studies using focus group discussions and semi-structured interviews. Results: The findings from previous studies in Dikgale HDSS revealed that an epidemiological transition is occurring. Again, the most widely reported barriers from previous studies in this rural area were: lack of knowledge of NCDs; shortages of medication and shortages of nurses in the clinics, which results in patients having long waiting-time at clinics. Lack of training of health care providers on the management of chronic diseases and the lack of supervision by the district and provincial health managers, together with poor dissemination of guidelines, were contributing factors to the lack of knowledge of non-communicable diseases (NCDs) management among nurses and community health care workers (CHWs). Consideration of all of these findings led to the development of model which focuses on integrating nursing services, CHWs and traditional health practitioners (THPs), including a well-established clinical information system for health care providers. A novel aspect of the model is the inclusion of community ambassadors who are on treatment for NCDs and are, thus, repositories of knowledge who can serve as a bridge between health care workers and community members. Conclusion: The model developed highlights the need for health interventions that aim to control risk factors at the population level, the need for availability of NCD-trained nurses, functional equipment and medication and a need to improve the link with traditional healers.
# Background
Chronic health conditions represent a substantial challenge to global health and by 2020 they will account for 73% of all deaths, constituting 60% of the global burden of disease [bib_ref] Preventing chronic diseases: taking stepwise action, Epping-Jordan [/bib_ref]. In South Africa NCDs have increased over the past 15 years. They now account for an estimated 37% of all-cause mortality and 16% of disability-adjusted life years. The overall level of NCD mortality is similar across all provinces in South Africa but the causes are different. Most NCDs share risk factors [bib_ref] Risk factors for chronic disease in Viet Nam: a review of the..., Hoy [/bib_ref] , and many of them are modifiable [bib_ref] Global response to non-communicable disease, Cerqueira [/bib_ref] and are usually adopted early in life. This provides considerable opportunities for intervention [bib_ref] Global response to non-communicable disease, Cerqueira [/bib_ref] ; however, progress in reducing NCD risk factors will only be attained if appropriate attention is given to the social and cultural contributing factors.
Chronic disease management (CDM) entails an integrated approach involving the patient, the family and the community as active participants over a lifetime of care [bib_ref] Diabetes care in sub-Saharan Africa, Beran [/bib_ref] [bib_ref] Integrated care for chronic conditions: the contribution of the ICCC framework, Nuňo [/bib_ref]. Effective interventions to limit the progression of the diseases or to mitigate the risk of complications are needed [bib_ref] Cardiovascular diseases and Diabetes as economic and developmental challenges in Africa, Kengne [/bib_ref]. Most CDM interventions should be tailored to the person's strengths and challenges in managing their care [bib_ref] Patientcentered care in chronic disease management: a thematic analysis of the literature..., Hudon [/bib_ref]. Thus, collaborative efforts on the management of people with chronic diseases in primary health care (PHC) settings plays an important role in CDM, as PHC professionals link their services to other specialised services in the communities [bib_ref] Chronic disease management in primary care: from evidence to policy, Dennis [/bib_ref]. Therefore, PHC requires a greater level of organisation that must be sustained, commonly over a patient's lifetime [bib_ref] Improving the prevention and management of chronic disease in lowincome and middle-income..., Beaglehole [/bib_ref]. Developing these interventions is a challenge in low-income and middle-income countries [bib_ref] Improving the prevention and management of chronic disease in lowincome and middle-income..., Beaglehole [/bib_ref]. The aim of this study was to develop an integrated, evidence-based CDM model based on the premise that multiple-strategy interventions are consistently more effective than single-strategy interventions [bib_ref] A critical review of interventions to increase compliance with medication-taking, obtaining medication..., Newell [/bib_ref].
# Methods
The research methodology for the current study was guided by the research question, which was developed in order to understand the burden of chronic disease risk factors; and to learn about the perceptions, experiences, barriers and challenges of chronic disease with respect to patients, nurses, CHWs and THPs regarding chronic disease management. This, therefore, brought about the idea of employing a mixed methods approach [bib_ref] Designing a mixed methods study in primary care, Creswell [/bib_ref] with the aim of integrating quantitative and qualitative data collection and analysis into a single study in order to develop an intervention program in a form of a model designed to improve the management of chronic diseases in a rural area. The use of this mixed-method methodology included the principle of making a decision on the priority or weight given to the quantitative and qualitative data collection and analysis in the study; the sequence of the data collection and analysis; and, the stage/stages in the research process at which the quantitative and qualitative data are connected and the results are integrated. When used in combination, both quantitative and qualitative data yielded a more complete analysis and they complemented each other, bringing about an understanding of the strengths and weaknesses of quantitative and qualitative research methodologies.
Multiple data was collected using different strategies, approaches and methods [bib_ref] Mixed methods research: a research paradigm whose time has come, Johnson [/bib_ref]. The quantitative techniques used included a cross-sectional study design using the World Health Organization (WHO) STEPwise approach to surveillance of NCD risk factors (WHO STEPS) [bib_ref] Preventable risk factors for noncommunicable diseases in rural Indonesia: prevalence study using..., Ng [/bib_ref]. The qualitative techniques used included focus group discussions (FGDs) [bib_ref] A proposed model for the analysis and interpretation of focus groups in..., Massey [/bib_ref] , semi-structured interviewsand QCs [bib_ref] Can quality circles improve hospital-acquired infection control?, Forster [/bib_ref] [bib_ref] The effect of quality circles on job satisfaction and quality of work-life..., Hosseinabadi [/bib_ref]. The present study was developmental in nature because the first method was used sequentially to help inform the second and third methods. This was done in the form of sequential triangulation because the model or interventions were developed in phases, with results from all the phases essential for planning the model development.
A multifaceted intervention using QCs [bib_ref] Teamwork in services: quality circles by another name?, Boaden [/bib_ref] [bib_ref] Quality circles to improve prescribing patterns in primary medical care: what is..., Wensing [/bib_ref] [bib_ref] The development of quality circles/peer review groups as a method of quality..., Beyer [/bib_ref] , was developed during the month of February 2014 to improve the quality and the management of chronic diseases in Dikgale HDSS. This intervention comprised small group sessions of expertise to discuss and provide feedback on prevention, management and control of chronic diseases in rural areas. The QCs have become an important method of quality improvement (QI) in primary care [bib_ref] The development of quality circles/peer review groups as a method of quality..., Beyer [/bib_ref]. The rationale behind the use of this research methodology was the fact that it involved gathering together small groups of employees doing similar or related work, who voluntary meet to identify, define, analyze and solve work-related problems or issues.
## Study setting and sampling
The study was conducted in Dikgale, a rural area which has a HDSS consisting of 15 villages, with poor infrastructure, situated close to one another in the Capricorn District of the Limpopo Province in South Africa [bib_ref] Spatial and temporal clustering of mortality in Digkale HDSS in rural northern..., Kanjala [/bib_ref]. The organisation of health services for chronic diseases in the study area is based on primary health care. These services are not complemented by more specialised and intensive care settings, such as diagnostic labs, specialty care clinics, hospitals and rehabilitation centres [bib_ref] The perceptions and perspectives of patients and health care providers on chronic..., Maimela [/bib_ref].
## Participants of quality circles
The QCs comprised of members representing a wide area of expertise with respect to disease control and prevention. Eligibility criteria included people who work as clinic managers in the study area, people with bachelor's degrees in nursing or associate degrees, people who have conducted extensive research in chronic disease management, an executive manager in Department of Health in Limpopo and a chronic disease manager in the Limpopo Province. An invitation was sent to several institutions employing people with expertise in chronic disease management and at total of 35 participants (as presented in [fig_ref] Table 1: Participants in the quality circles Staff and students from University of Limpopo... [/fig_ref] were randomly selected from this group to take part in the study.
Patients, nurses, community health workers and traditional health practitioners were excluded in this study because of the use of the QCs methodology [bib_ref] Community interventions for health (CIH): a novel approach to tackling the worldwide..., Duffany [/bib_ref] , which was seen as an innovative approach to addressing quality improvement activities. However, the responses from this group of people elicited in previous studies, which formed part of the major project, were used within the context of CDM improvement in PHC settings. This process involved the identification of, and discussion about, actual problems faced when addressing chronic diseases and possible solutions to these problems, which were raised by patients, nurses, CHWs, THPs and managers in earlier studies in the same study area.
Next, priorities for an integrated community-based chronic disease approach were set and a start was made on an implementation plan.
## Data collection and analysis
This study took the form of a workshop, which was conducted over three days, in order to discuss problems related to the improvement of CDM. Three small groups of approximately 11 participants were formed and several educational/presentational sessions were held to familiarise the participants with the issues of group dynamics, problem-solving, techniques used in QCs, the role of each member of the group and the procedures used in the QCs. These small groups were tasked to discuss current practices and explore evidence-based practices [bib_ref] Teamwork in services: quality circles by another name?, Boaden [/bib_ref] required to improve CDM in the study area. The groups worked through a series of stages, during which participants were encouraged to analyse problems following a sequential process, in order to find possible causes of these problems and then develop solutions. Groups then formally presented project proposals for consideration [bib_ref] Revisiting cyberbullying in schools using the quality circle approach, Paul [/bib_ref].
Interventions which formed the integrated CDM model were discussed, taking into account the evidencebased and/or grounded theory aspects of each intervention. The basis for the discussion was the findings from previous studies [bib_ref] The perceptions and perspectives of patients and health care providers on chronic..., Maimela [/bib_ref] , which formed part of the larger project in the study area, i.e. the STEPwise survey, focus group discussions and semi-structured interviews. The STEPwise survey results dealt with the prevalence of risk factors for NCDs. The focus group discussions results dealt with the perceptions and perspectives of chronic patients and health care providers on CDM [bib_ref] The perceptions and perspectives of patients and health care providers on chronic..., Maimela [/bib_ref] , addressing NCDs through CHWs and THPs. The semistructures interviews results dealt with the barriers to, and facilitators for, improving the management of chronic NCDs at PHC level by managers of clinics; and the CHWs programme and chronic disease programme at district and provincial level. Model development was also based upon an assessment of the needs, which were inclusive of consultation with key stakeholders and involved a multidisciplinary approach, where applicable, which considered the optimal and equitable utilisation of healthcare resources [bib_ref] Quality circles and their potential application to rural health care in Papua..., Cibulskis [/bib_ref].
# Results
In the previous studies conducted as part of the bigger project in the study area, the quantitative study revealed a high prevalence of behavioural and biomedical risk factors for NCDs [bib_ref] The prevalence and determinants of chronic non-communicable disease risk factors amongst adults..., Maimela [/bib_ref]. Approximately one in three study participants were found to be hypertensive, starting at a young age of 15 to 24 years. Approximately 90% of the participants were below the WHO recommendations for fruit and vegetable consumption, while Expertise from University of Umeå, Sweden more than half had low physical activity levels. A quarter of the participants were overweight and obese, while one in third had high total cholesterol levels. The qualitative studies [bib_ref] The perceptions and perspectives of patients and health care providers on chronic..., Maimela [/bib_ref] revealed that lack of knowledge of chronic diseases was predominant among patients, nurses, CHWs and THPs. Training on CDM was also lacking or insufficient and there was poor supervision of health facility operations by the district and provincial managers. The CHWs were not respected by nurses and their remuneration was not regularly received. A poor relationship was found to exist between THPs and clinic nurses, due to lack of formal referral system. This was primarily due to the lack of a formal structure to represent the THPs in government. The QCs used the abovementioned findings to discuss specific needs in relation to capacities for chronic disease prevention and management in the Dikgale HDSS. The discussions were focused on the ability to integrate determinants of health approach into programme planning in order to address the root causes of chronic disease. Further discussions addressed the integration of health services in order to improve the management and prevention of chronic non-communicable diseases. Therefore, the model development followed three major steps as follows:
Firstly, to identify inputs (resources), activities, outputs and outcomes in a form of logic framework to improve management of chronic NCDs; Secondly, to identify the prerequisites needed to strengthen integrated evidence-based chronic NCDs; Lastly, to develop an integrated evidence-based chronic NCD management model.
Logical framework for improvement of management and prevention of chronic non-communicable diseases A framework in the form of a logic model was established from the results in order to systematically lay out the programme elements and a path showing what could be done to improve management and prevention of chronic NCDs. The logic framework is presented in . The main activities in the framework focus on capacity building for health care providers; community screening and referral of chronic patients to clinics; and, referral of patients back to CHWs and THPs for monitoring in the community. The outputs required to attain the desired outcomes are trained health care providers, counselled community members, screened community members and referred patients. The outcome measures for these interventions would be improved quality of NCD management; increased access to NCD screening; increased knowledge about risk factors for NCDs; and, access to treatment and preventions strategies in the communities.
The intermediate expected outcomes as a result of the impact of the interventions will likely be a decrease in risk behaviour and an increase in access to NCD treatment. Decreased NCD risk factors, increased NCD incidences Logic Model to improve management and prevention of chronic non-communicable diseases and decreased NCD morbidity and mortality will be assessed as the impact parameters of the interventions. The interventions in the proposed framework can strengthen the development of an integrated evidencebased chronic NCD management model, focusing on its implementation and later evaluation. This framework will be used to organise thinking around the model development; how to relate model activities and investment on expected results, which are improved chronic noncommunicable disease management. The framework will also be used to set up performance indicators in the clinics; allocating responsibilities to all involved in chronic care and, finally, communicating information on the model implementation concisely and unambiguously.
## Inputs and activities to strengthen integrated evidencebased chronic non-communicable disease management model at dikgale hdss
The discussions emanating from the QCs and executive management of the Limpopo Department of Health, together with Belgian Vlaamse Interuniversitaire Raad (VLIR) team, resulted in inputs and activities which are needed for the development and implementation of the interventions for improving CDM. Looking at the current health system in the Limpopo Province, the team suggested that the interventions should optimally and equitably utilise the current available healthcare resources to implement an integrated model which is culturally sensitive and appropriate to the population of the Dikgale area. The inputs and activities agreed upon to guide in the process to achieve improved chronic NCD management are outlined schematically in
## As follows:
There are prerequisites which were identified from the qualitative results that can be implemented to enhance collaboration among health care practitioners, such as addressing their attitudes. This can nurture the working relationship among nurses, CHWs and THPs. Again the provision or availability of Standard Operation Procedures (SOPs) for health care practitioners was emphasised in order to standardise quality health care services by nurses, CHWs and THPs. The creation of a supportive environment by the district and provincial offices for the health care workers was identified as a need to enhance the environment for all employees to improve their productivity and morale. Health care practitioners' readiness can be improved by conducting training, which can impart knowledge Prerequisites needed to strengthen integrated evidence-based chronic non-communicable disease management Dikgale HDSS to them and prepare them. Therefore, the strengthening of collaboration and integration of health care practitioner-services to serve the poor communities in the Dikgale area should be emphasised. The activities addressing periodic community screening targeting high risk groups and the raising of awareness on common NCDs and their risk factors in the community were identified. The findings from the qualitative study show that awareness campaigns should use approaches, such as community radio stations, community dialogue and mass campaigns, to reach more community members. Establishment of a surveillance system to monitor the occurrence of NCDs and their risk factors was also identified as an intervention needed in the health facilities. The involvement of the people themselves, family and community members, was seen as a critical aspect which can be done through the establishment of a chronic NCD management and health promotion forum. This forum should collaborate with community members to establish the chronic disease ambassador programme to will motivate patients in the communities through their health problems.
## Integrated evidence-based chronic non-communicable disease management model in dikgale hdss
The logical framework , together with the activities and inputs , presented above led to the development of an integrated chronic non-communicable disease management model, which was the main objective of this study, and is presented in [fig_ref] Figure 3: Integrated, evidence-based chronic non-communicable disease management model in the Dikgale HDSS [/fig_ref]. As this model was developed and proposed based on the findings from the study, it is therefore evidence-based, i.e. based on the local situation in the Dikgale HDSS. It was developed in an integrated or coordinated approach to service delivery. This model describes four interacting system components, namely, health care providers, the health care system, community partners and patients with their families. The main feature of this model is the integration of services provided by nurses, CHWs and THPs. A wellestablished clinical information system is proposed in the model for health care providers to have access to more informed patient care. The health care system based on this model will be tasked to organise health care in the rural area to improve management and prevention of chronic illnesses. Support systems in a form of supervisory visits to clinics, provision of medical equipment and the training of health care providers should be provided. A contribution from community partners, in the form of better leadership to mobilise and coordinate resources for chronic care, is emphasised in the model. This productive interaction will be supported by the district and provincial Health Departments through re-organisation of health services to give traditional leaders a role to participate in leadership and to improve community participation.
This study suggests patients may be used as ambassadors, informing others and will play a role to motivate other patients in the communities through their health problems. As the ambassadors will be drawn from those community members who are suffering from chronic diseases themselves but have managed their chronic illness well in the past, it will be good for chronic patients to learn from the best practices of those who managed their conditions well in the past. They will be tasked to encourage chronic patients on the self-management of their condition, together with the involvement of their family members and community partners at large. Informed patients who will form part of the prepared and proactive ambassador team will contribute to functional and clinical outcomes in the community. This will eventually lead to improved health outcomes, reduced burden of chronic diseases and improved sustainability of health system.
# Discussion
The integrated evidence-based chronic NCD management model developed in the current study describes changes which are needed at PHC settings in a rural area.
The main focus of our study was to develop appropriately effective intervention strategies that will facilitate change, in order to prevent and control chronic diseases within and across cultural contexts [bib_ref] Community interventions for health (CIH): a novel approach to tackling the worldwide..., Duffany [/bib_ref] in rural areas. An integrated multi-faceted approach to improving care of chronic conditions has been shown to result in better outcomes [bib_ref] Implementing the chronic care model for improvements in diabetes practice and outcomes..., Siminerio [/bib_ref].
These changes are based on four interacting system components, which are health care providers; health care system; community partners; and, patients with their families. These health system changes are supported by the concept of an ideal clinic introduced into the South African health system, in order to move from a hospital-centric, curative system to preventive and promotive PHC that is cost effective and meets community needs. This is also supported by the Chronic Care model developed by Wagner and colleagues, which comprises four components, which are self-management support; delivery system design; decision support; and clinical information systems [bib_ref] Preventing chronic diseases: taking stepwise action, Epping-Jordan [/bib_ref]. An important difference in the current proposed integrated model is the integration of services from nurses, CHWs and THPs. This addition to the model was mainly due to the fact that, in this rural area, health services are provided by nurses in the clinics and CHWs and THPs in the community. This study also supports the notion that patients may be the ambassadors who inform and support others, as seen in recent smoking cessation programmes [bib_ref] Smoking and chronic obstructive pulmonary disease (COPD), Laniado-Laborin [/bib_ref] and also in decision making [bib_ref] A new taxonomy for stakeholder engagement in patient-centered outcomes research, Concannon [/bib_ref].
## Implications for health care and further research
There are two gaps which must be closed to achieve proper control of chronic diseases, which are the gap between effective interventions in research studies and what clinicians do in practice, and the gap between what clinicians in their offices recommend to patients and what patients do at home and in their communities [bib_ref] Evidence based medicine: what it is and what it isn't, Sackett [/bib_ref]. To successfully close these gaps, we have discussed the interventions with experts in CDM, including clinicians and researchers, together with the executive management of the Department of Health in Limpopo Province, as well as evidence collected from the health facilities in the study area. The public health implications from the study are that our interventions are oriented towards health promotion and prevention through a primary health care approach in order to effectively respond to the complex social, cultural and behavioural issues associated with NCDs.
# Conclusions
The model developed highlights the need for health interventions that aim to control risk factors at the population level, the need for availability of NCD-trained nurses, functional equipment and medication and a need to improve the link with traditional healers. Therefore, further research will be conducted in the study area on planned community intervention programmes, which are a substantial component of the strategy to help solve prevention of NCDs [bib_ref] Cardiovascular diseases and Diabetes as economic and developmental challenges in Africa, Kengne [/bib_ref]. Therefore, as the model development is comprehensive [bib_ref] Evidence based medicine: what it is and what it isn't, Sackett [/bib_ref] [bib_ref] An evaluation of collaborative interventions to improve chronic illness care. Framework and..., Cretin [/bib_ref] emphasis should also be placed on the complex and dynamic settings in which the delivery of nursing care occurs, as it involves social, political, economic and clinical factors [bib_ref] Beyond the rhetoric: what do we mean by a "model of care"?, Davidson [/bib_ref].
## Acknowledgements
We would like to thank Medical Science Department, University of Limpopo, in South Africa and International Health Unit, Antwerp University in Belgium. Similar and invaluable support is acknowledged from the team members who participated in the quality circles to discuss the interventions and the executive management of the Department of Health in Limpopo.
# Funding
The financial support for the conduct of the research and preparation of the article was provided by University of Limpopo and the Vlaamse Interuniversitaire Raad (VLIR). These funding helped in paying field workers who collected quantitative data which was used in previous studies, payment of tuition fees for the main authors PhD degree and lastly, contributed to the design of the study as there was a workshop arranged and paid by the funder to conduct the quality circles which helped in data analysis and report writing.
## Availability of data and materials
The data can be obtained by request from the corresponding author.
Authors' contributions EM made substantial contributions to conception, design, acquisition of data, analysis, interpretation of data and drafting the manuscript including revising it critically for important intellectual content. MA made substantial contributions into the design of the study and revising it critically for important intellectual content. HB supported data-analysis, interpretation of data and contributed to the drafting of the manuscript. JF supported data-analysis, interpretation of data and contributed to the drafting of the manuscript. HM made substantial contributions into the design of the study and revising it critically for important intellectual content. JW supported data-analysis, interpretation of data and contributed to the drafting of the manuscript. JP made substantial contributions into the design of the study and revising it critically for important intellectual content. All authors provided input into drafts and approved the final draft of the article.
Ethics approval and consent to participate Ethical approval to conduct the study was sought from the University of Limpopo Medunsa Research Committee which is a research ethics committee for University of Limpopo (MREC/HS/05/2013:PG). Institutional or departmental approval to conduct the study in health facilities was granted by the Provincial Research Committee which serves as the review board for the Department of Health in Limpopo Province. Written informed consent for participation was obtained from the participants and they were provided with information leaflet about the study and how should quality circles be conducted.
## Consent for publication
Not applicable.
Competing interests I hereby declare that there are no competing interests for all the authors involved in the pursuance of this study and that there are also no financial gains, as the interests were solely academic, aimed at contributing to an improvement in chronic disease management in rural areas of the Limpopo Province.
[fig] Figure 3: Integrated, evidence-based chronic non-communicable disease management model in the Dikgale HDSS [/fig]
[table] Table 1: Participants in the quality circles Staff and students from University of Limpopo Medical Science Department 10 Health Promotion from Limpopo Department of Health 2 Staff member from MRC/Wits-Agincourt Research Unit in Mpumalanga province 1 A representative from the executive management of Department of Health in Limpopo Province 2 A representative from the Dikgale traditional authority 1 Expertise from Antwerp University in Belgium representing the Unit International Health, Department of Primary and Interdisciplinary care, Department of Sociology and Research Methodology. [/table]
|
Antibiotic-Resistant Acinetobacter baumannii Increasing Success Remains a Challenge as a Nosocomial Pathogen
Antibiotic-resistant infectious bacteria currently imply a high risk and therefore constitute a strong challenge when treating patients in hospital settings. Characterization of these species and of particular strains is a priority for the establishment of diagnostic tests and preventive procedures. The relevance of Acinetobacter baumannii as a problematic microorganism in inpatient facilities, particularly intensive care units, has increased over time. This review aims to draw attention to (i) the historical emergence of carbapenem-resistant Acinetobacter baumannii, (ii) the current status of surveillance needs in Latin America, and (iii) recent data suggesting that A. baumannii continues to spread and evolve in hospital settings. First, we present synopsis of the series of events leading to the discovery and precise identification of this microorganism in hospital settings. Then key events in the acquisition of antibiotic-resistant genes by this microorganism are summarized, highlighting the race between new antibiotic generation and emergence of A. baumannii resistant strains. Here we review the historical development of this species as an infectious threat, the current state of its distribution, and antibiotic resistance characteristics, and we discuss future prospects for its control.
# Introduction
The first report of Acinetobacter infections in a hospital setting, within an intensive care unit (ICU), dates back to the 1960s [bib_ref] Analysis of hospital bacteriological data, Stirland [/bib_ref]. Phenotypic tests were initially used to ascribe gender to the organism causing these infections. Identification at the species level was only possible when molecular tests were employed. Difficulties in species identification may have caused delays in recognizing the microorganism and hence the impact of Acinetobacter baumannii as a nosocomial pathogen was underestimated. The persistent presence of A. baumannii in the hospital sector allowed it to come into contact with antibiotics. These environmental conditions exerted such a selective pressure that successful clones with particular antibiotic resistance characteristics emerged. Previous reviews have focused on A. baumannii as a successful pathogen, and its biological aspects, epidemiology, pathogenicity factors, and global spread have also been addressed. Here we review the increase of antibioticresistant Acinetobacter in recent reports and summarize the multilateral work being carried out in order to control this pathogen. As multinational networks have emerged based on the information provided by these reports, we update the available information and give an example of the importance of local reports. This review is organized as a timeline divided in three periods. The first period includes the decades of the 60s and 70s when the Acinetobacter gender started to be reported in hospitals. The second period includes the decades of the 80s and 90s during which baumannii species was determined and the reports of the organism's carbapenem resistance started. The last period includes years 2000 to 2015, during which World Health Organization (WHO) has issued calls for containment of antimicrobial resistance.
2 Journal of Pathogens haemolysans, Mima polymorpha, and Moraxella lwoffii [bib_ref] Moraxella, acinetobacter, and the mimeae, Henriksen [/bib_ref]. The current designation of the genus Acinetobacter (from the Greek akinetos, meaning nonmotile) was proposed in 1954 by Prevot and it was accepted in 1968 [bib_ref] Studies on bacterial taxonomy. X. The revision of species under Acromobacter group, Brisou [/bib_ref]. Finally, in 1974, this designation was included in Bergey's Manual of Systematic Bacteriology, where it was described as having only one species: Acinetobacter calcoaceticus [bib_ref] Taxonomy of Moraxellaceae fam. nov., a new bacterial family to accommodate the..., Rossau [/bib_ref]. Later, in 1986, Bouvet and Grimont observed inconsistencies in the application of phenotypic tests to determine the genus and species of Acinetobacter. These inconsistencies arose due to the fact that the members of this genus have different catabolic pathways that allow them to adapt to most substrates [bib_ref] Sequencing of the rpoB gene and flanking spacers for molecular identification of..., Scola [/bib_ref]. DNA hybridization studies were then introduced to determine the different species, and the homology groups with more than 70% DNA-DNA relatedness were called genomic species. Currently, there are 32 genospecies known. The complex A. calcoaceticus-baumannii includes four genospecies: genospecies 1, A. calcoaceticus; genospecies 2, A. baumannii; genospecies 3, A. pittii; and genospecies 13TU, A. nosocomialis. Among these, A. baumannii is the most important in the clinical context [bib_ref] Acinetobacter baumannii: emergence of a successful pathogen, Peleg [/bib_ref] [bib_ref] Acinetobacter calcoaceticus: a nosocomial pathogen with an unusual seasonal pattern, Retailliau [/bib_ref] , since it is the most frequently isolated in nosocomial infections and also the one associated with the highest mortality rate [bib_ref] Differences in carbapenem resistance genes among Acinetobacter baumannii, Acinetobacter genospecies 3 and..., Lin [/bib_ref] [bib_ref] Influence of genospecies of Acinetobacter baumannii complex on clinical outcomes of patients..., Chuang [/bib_ref].
## Increasing impact of a. baumannii at a global scale
It is difficult to determine the presence of A. baumannii in nosocomial infections in reports before the year 1974. At that time, there was a specific designation for a single species: A. calcoaceticus. Moreover, the routine laboratory diagnostic identification was carried out by phenotypic methods that are not useful for species designation. Particularly, these methods do not allow a distinction within the calcoaceticus-baumannii complex. Here we use identification and antibiotic resistance data arising from hospital settings where the pathogen has become successful.
A key question to understand the origin and impact of A. baumannii in hospital settings is the following: How did it become a pathogen? Olson has pointed out at least three contributing elements. First, water is an important reservoir for clinically significant Gram-Negative Nonfermenting (GN-NF) bacteria. Second, this species is a waterrelated organism, and solutions as well as surgical, medical, and clinical instruments can harbor high levels of bacteria. With bacteria already present, debilitated patients became susceptible to infection by this opportunistic microorganism, giving rise to nosocomial outbreaks [bib_ref] The identification of gram-negative, nonfermentative bacteria from water: problems and alternative approaches..., Ward [/bib_ref]. Early Acinetobacter outbreaks were easily controlled with common antibiotics including -lactams and sulfonamides. However, these treatments became ineffective very soon, coincident with early reports of emerging microbial resistance to these antibiotics. Unfortunately, since the 1970s, microbial resistance has only increased; at the same time, the creation of effective new antibiotics has declined [fig_ref] Figure 1: Top diagram shows dates of introduction of antimicrobials [/fig_ref]. Emergence of such antibiotic-resistant strains has made the management of patients and the containment and control of outbreaks more difficult. These difficulties also increase morbidity, mortality, and hospital costs, making these outbreaks a challenge for public health systems [bib_ref] An enzyme from bacteria able to destroy penicillin, Abraham [/bib_ref]. The emergence of A. baumannii as a threat in hospital settings has common historical developments in diverse geographical locations worldwide. One of the most relevant events, besides the refinement of laboratory techniques to identify this species, is the acquisition of antibiotic resistance.
## Emergence and establishment of
A. baumannii as a Nosocomial Microorganism: 1960s and 1970s
At this period in time, A. baumannii emerged as a microorganism causative of nosocomial infections that were easily managed. Successful treatment of those nosocomial infections was achieved simply with -lactams. These reports arose mainly from hospitals located in Europe and the United States. The earliest reports of infections caused by this organism in a hospital setting came from Europe. Computer analysis of bacteriological data collected in 1965 in a German hospital revealed the presence of the Acinetobacter genera. In this case, A. anitratus was found to be present in blood culture samples, and no antimicrobial resistance was reported. By the end of the 1970s, some other countries reported the pathogen as a cause of infection outbreaks, also detected in blood culture samples [bib_ref] Analysis of hospital bacteriological data, Stirland [/bib_ref] [bib_ref] Acinetobacter calcoaceticus outbreak associated with peritoneal dialysis, Abrutyn [/bib_ref]. -lactams and sulfonamides were used as an effective treatment for these infections up to 1975. That year, A. calcoaceticus was found to be the cause of an outbreak, despite the fact that it has been considered a low pathogenic microorganism. The microorganism causing this hospital epidemic showed a noticeable reduction in susceptibility to those antibiotics, thus explaining its higher impact [bib_ref] A hospital epidemic due to Achromobacter calcoaceticus, Lecocq [/bib_ref]. Both in the United States and in Europe, earlier reports also date back to 1965. That year, a review on the Moraxella group included a report where Acinetobacter was described as belonging to the Gram-negative bacilli causing pneumonia [bib_ref] Changing pharyngeal bacterial flora of hospitalized patients-emergence of gram-negative bacilli, Johanson [/bib_ref] [bib_ref] The Moraxella lwoffi group of bacteria; a review, Henderson [/bib_ref]. Latin American countries did not take long to begin reporting findings of this organism as a cause of nosocomial infections, starting in 1979 [bib_ref] Acinetobacter calcoaceticus: a nosocomial pathogen with an unusual seasonal pattern, Retailliau [/bib_ref] [bib_ref] Multiple intensive care unit outbreak of Acinetobacter calcoaceticus subspecies anitratus respiratory infection..., Hartstein [/bib_ref]. Unfortunately, by the end of the 1970s, Acinetobacter was reported to be resistant to sulfonamides, -lactams, and aminoglycosides. These antibiotics were previously considered the treatment of choice, but by this time they were no longer effective [bib_ref] Antimicrobial resistance of Acinetobacter spp, Van Looveren [/bib_ref]. Thus, in just about two decades, Acinetobacter became consolidated as a pathogen causing hospital outbreaks difficult to treat and contain on both sides of the Atlantic.
## Outbreak increase and cumulative antibiotic resistance: 1980s and 1990s
This period was characterized by the specific designation of the genetic species within the complex and by the critical identification of A. baumannii species. In fact, it was only until 1986 that it became possible to attribute Acinetobacter infections specifically to A. baumannii. At the same time, outbreak analysis studies revealed increasing antimicrobial resistance [fig_ref] Figure 1: Top diagram shows dates of introduction of antimicrobials [/fig_ref]. The cumulative nature of antimicrobial resistance posed a new problem for the control and containment of these outbreaks. By the end of the 1990s, Asia and Latin America had integrated reports of outbreaks caused by antibioticresistant A. baumannii strains to their data. At this time, the geographical spread of outbreaks led to a change in the surveillance and in the levels of reporting. It became relevant to maintain both a regional surveillance and a worldwide surveillance of the outbreaks. This was carried out with a particular focus on the characterization of antibiotic resistance patterns based on molecular techniques. In order to find ways to contain these outbreaks, European countries started the search for risk factors. These studies led to the establishment of an association between ventilator use and increased mortality with strains that were resistant to more than three antibiotic groups [bib_ref] Risk factors for nosocomial colonization with multiresistant Acinetobacter baumannii, Mulin [/bib_ref] [bib_ref] Nosocomial pneumonia in patients receiving continuous mechanical ventilation: prospective analysis of 52..., Fagon [/bib_ref] [bib_ref] Progressive resistance in a single strain of Acinetobacter calcoaceticus recovered during a..., Carlquist [/bib_ref]. These strains were therefore designated as multidrugresistant (MDR) microorganisms. At this point, carbapenems had emerged, and in 1985 they were used as the therapeutic option to treat infections by MDR bacteria. But unfortunately, despite the implementation of the imipenem treatment, resistance to these antibiotics was reported in this same year. Surprisingly, the enzymes responsible for such antibiotic resistance were the OXA--lactamases, already widespread among infectious microorganisms [bib_ref] Carbapenem resistance mediated by beta-lactamases in clinical isolates of Acinetobacter baumannii in..., López-Hernández [/bib_ref]. Those findings led to the implementation of association studies based on the phenotypic determination of the antibiotic resistance of outbreak isolates. A variety of methods were then applied to typify and compare hospital outbreak isolates as well as community isolates. These studies became necessary in order to generate new strategies to control the outbreaks [bib_ref] Comparison of outbreak and nonoutbreak Acinetobacter baumannii strains by genotypic and phenotypic..., Dijkshoorn [/bib_ref] [bib_ref] Multicenter study using standardized protocols and reagents for evaluation of reproducibility of..., Grundmann [/bib_ref] [bib_ref] Correlation of typing methods for Acinetobacter isolates from hospital outbreaks, Dijkshoorn [/bib_ref] [bib_ref] Correlation of six methods for typing nosocomial isolates of Acinetobacter baumannii, Marcos [/bib_ref]. The increasing number of cases associated with Acinetobacter in the United States made it necessary to establish a set of specific features associated with the Acinetobacter infections. This in turn led to the finding of an association with a diversity of risk factors, including ventilator use and previous antimicrobial therapy [bib_ref] An outbreak of nosocomial acinetobacter infections from humidifiers, Gervich [/bib_ref] [bib_ref] An outbreak of acinetobacter infection associated with the use of a ventilator..., Irwin [/bib_ref]. An important observation was that the number of isolates was proportional to the increase in the number of antibiotics to which the microorganism was resistant. In general, the increased frequency of A. baumannii in clinical isolates correlated with a progressive increase in antibiotic resistance [bib_ref] Progressive resistance in a single strain of Acinetobacter calcoaceticus recovered during a..., Carlquist [/bib_ref].
A key event in the characterization of A. baumannii's impact as a microorganism causative of nosocomial infections was the species designation in 1986. This led to the realization that most infections attributed to Acinetobacter were in fact caused by baumannii species. It was then assumed that there was a single Acinetobacter species of clinical interest, and several biotypes were reported. However, the phenotypic methods used for species typing were producing inconsistent results. Therefore, molecular methods, such as plasmid characterization, were introduced in order to analyze the antibiotic resistance mechanisms. By the end of the 1980s, nosocomial reports characterizing A. baumannii infections had notably increased, and special focus on antibiotic resistance features uncovered then reduced susceptibility to imipenem [bib_ref] Prevalence of Acinetobacter spp. isolates with reduced susceptibility to imipenem, as determined..., Jones [/bib_ref] [bib_ref] Acinetobacter, Stratton [/bib_ref] [bib_ref] Epidemic bacteremia due to Acinetobacter baumannii in five intensive care units, Beck-Sagué [/bib_ref]. At this period in time, there was also an increase in the reports of Acinetobacter-positive nosocomial isolates in Latin America. Despite the limitations to the identification, the first half of the 1990s produced the highest number of reports, although the pathogen was still being described as A. calcoaceticus variety anitratus. These findings led to the introduction of better hygiene practices including thorough hand washing to prevent transmission within hospital settings [bib_ref] Intrahospital septicemia due to Acinetobacter calcoaceticus var, Vivanco [/bib_ref] [bib_ref] Effectiveness of hand-cleansing agents for removing Acinetobacter baumannii strain from contaminated hands, Cardoso [/bib_ref] [bib_ref] Bacteriology of middle ear fluid specimens obtained by tympanocentesis from 111 Colombian..., Trujillo [/bib_ref] [bib_ref] Acinetobacter calcoaceticus anitratus outbreak in the intensive care unit traced to a..., Ahmed [/bib_ref]. It was only after 1990 that Asia started to produce reports of A. baumannii's role in nosocomial infection outbreaks. Before 1990, reports from Asia were related only to environmental food-borne Acinetobacter strains. The first A. baumannii strain from a hospital setting had low pathogenicity. However, by 1994, multiresistant strains were being detected in Asia, and they became a great public health challenge, just as it was happening in other continents [bib_ref] The epidemiology of Acinetobacter infections in Hong Kong, Siau [/bib_ref] [bib_ref] Antimicrobial susceptibilities of clinical isolates of Acinetobacter baumannii from Singapore, Kuah [/bib_ref].
## Creation of surveillance networks: 2000 to 2015
The most recent period has been characterized by the dissemination of carbapenem-resistant A. baumannii (CRAB) strains and the creation of international surveillance networks [bib_ref] Acinetobacter baumannii: emergence of a successful pathogen, Peleg [/bib_ref]. This has led to the implementation of measures to achieve the containment of outbreaks of this antibioticresistant pathogen, which has turned into a global priority. The first international call for containment was issued in 2001. WHO proposed a series of recommendations to slow down the emergence and reduce the spread of bacterial resistance; they were described in "Global Strategy for Containment of Antimicrobial Resistance," published that year. This document considers that the surveillance of antimicrobial resistance should be carried out among common pathogens in the community as well as in hospital and other healthcare facilities. It is then recommended that surveillance should be directed to microorganisms with high antibiotic resistance index. This group of microorganisms is designated by the acronym ESKAPE, and it includes Acinetobacter. The worldwide presence of antibiotic-resistant microbes has made surveillance shift from a local scale to a global one [fig_ref] Figure 2: Brief history of the incorporation of Acinetobacter baumannii as one of the... [/fig_ref]. Therefore, the challenge that emerges is to contain the dissemination of the resistant strains, leading to the formation of multinational surveillance networks. The first was the European Antimicrobial Resistance Surveillance System (EARSS) constituted in 1998 and renamed as European Antimicrobial Resistance Surveillance Network (EARS-Net) in 2010. This network is coordinated by the European Centre for Disease Prevention and Control (ECDC). The 2013 Annual Epidemiological Report showed a frequency of CRAB strains of 3.6% in healthcare-associated infections; 81.2% were carbapenem nonsusceptible.
In the United States, the national public health surveillance system is a collaborative project between CDC, FDA, USDA, and state and local health departments. The CDC together with the Interagency Task Force on Antimicrobial Resistance (ITFAR) coordinates different agencies for the containment of resistance. The 2013 CDC report shows that Acinetobacter was the cause of 2% of nosocomial infections that year. However, in critically ill patients on mechanical ventilators, it was responsible for about 7% of the infection cases. Moreover, 7,300 (63%) out of 12,000 annual Acinetobacter infections were MDR, which in turn has led to about 500 annual deaths attributed to these infections. Acknowledging that the growing problem of bacterial resistance is a public health challenge, the USA (CDC) and the EU (ECDC) established the Transatlantic Taskforce on Antimicrobial Resistance (TAFTAR) in 2009. This network was created to foster international collaboration on the prevention and control of antimicrobial resistance (http://www.cdc.gov/drugresistance/tatfar/about.html). The highest levels of bacterial resistance to carbapenems were reported in Latin American countries. These findings resulted in the implementation of antimicrobial resistance surveillance at a national level in Brazil, Argentina, and Colombia, with special focus on uncovering molecular mechanisms [bib_ref] Outbreak of carbapenem-resistant Acinetobacter baumannii producing the OXA-23 enzyme in Curitiba, Brazil, Dalia-Costa [/bib_ref] [bib_ref] Identification of an epidemic carbapenem-resistant Acinetobacter baumannii strain at hospitals in Buenos..., Barbolla [/bib_ref]. Additional studies provided by the SENTRY Antimicrobial Surveillance Program and other programs have reported different rates of resistance in Latin American countries. The highest resistance to imipenem was detected in Argentina (20%), followed by Colombia (14%) and Brazil (8.5%). The increase in MDR A. baumannii was noticed for the first time in 2001 by Gales and collaborators [bib_ref] Emerging importance of multidrug-resistant Acinetobacter species and Stenotrophomonas maltophilia as pathogens in..., Gales [/bib_ref].
Despite a general increase of carbapenem-resistant Acinetobacter isolates, specific reports have rarely arisen from Chilean agencies. In Brazil, apparently decreasing rates were reported between 1997 and 2011 (13.6% to 2.2%) [bib_ref] Resistance trends of Acinetobacter spp. in Latin America and characterization of international..., Tognim [/bib_ref]. However, a study revealed a real increase in imipenem resistance: from 12.6% for the 1996-1997 period to 71.4% for the 2008-2010 period. Increases in Argentina and Chile were 84.9% and 50%, respectively [bib_ref] Antimicrobial resistance among Gram-negative bacilli isolated from Latin America: results from SENTRY..., Gales [/bib_ref]. In addition to the national surveillance systems, the Latin American Antimicrobial Resistance Surveillance Network (ReLAVRA-PAHO) was created. It specified that efforts should be made to carry out Acinetobacter surveillance in hospital settings [bib_ref] Informe Anual de la Red de Monitoreo/Vigilancia de la Resistencia a los..., Ops-Usaid [/bib_ref] Within this network, each country has its own system and not all the countries report the resistance profile of nosocomial bacteria.
Currently, some countries in Asia are carrying out surveillance at a national level. These include Malaysia, Thailand, Pakistan, India, and Taiwan, where A. baumannii is one of the most common pathogens [bib_ref] Epidemiology, etiology, and diagnosis of hospitalacquired pneumonia and ventilator-associated pneumonia in Asian..., Chawla [/bib_ref]. In 2011, this microorganism was considered to belong to the group of pathogens characterized by carbapenem resistance [bib_ref] High prevalence of multidrug-resistant nonfermenters in hospital-acquired pneumonia in Asia, Chung [/bib_ref]. The existence of the new carbapenemase NDM-1 was reported in 2008. This enzyme originated in India and was quickly disseminated among bacteria, including A. baumannii [bib_ref] Emergence of NDM-1-producing Acinetobacter baumannii in China, Chen [/bib_ref]. International studies have shown that Asia contains a large number of resistant Gram-negative bacteria (GNB) [bib_ref] International Nosocomial Infection Control Consortium (INICC) report, data summary of 36 countries, Rosenthal [/bib_ref] [bib_ref] High prevalence of multidrug-resistant nonfermenters in hospital-acquired pneumonia in Asia, Chung [/bib_ref]. However, most Asian countries are currently lacking antimicrobial-resistant bacteria surveillance systems [bib_ref] The global spread of healthcare-associated multidrugresistant bacteria: a perspective from Asia, Molton [/bib_ref].
The data collected by international networks have allowed researchers to generate a global picture of the circulating antibiotic-resistant strains. Recent studies have shown the presence of three successful clones called ribogroups or World Wide Types distributed in some countries: WWI, WWII, and WWIII [bib_ref] Identification of a new geographically widespread multiresistant Acinetobacter baumannii clone from European..., Van Dessel [/bib_ref]. International agreements to standardize methodologies and allow wide data comparison have led to the implementation of multilocus sequence typing (MLST). This analysis includes the characterization of seven housekeeping genes to define a sequence type (ST). The associations of these sequence types, where there is similarity among six to seven alleles, were defined as clonal complexes. For instance, it has been shown that clonal complex CC92, previously known as WWII, is the most distributed one around the world [bib_ref] Multilocus sequence typing: a portable approach to the identification of clones within..., Maiden [/bib_ref] , followed by CC109 and CC87 (determined by the Oxford system). There are also international clones that can be found in a single continent: such is the case of CC636 (CC113) in Latin America [bib_ref] Insights into the global molecular epidemiology of carbapenem non-susceptible clones of Acinetobacter..., Karah [/bib_ref].
The beginning of the 21st century was marked by a sharp increase in international mobility. The intense travelling across countries and continents has had a strong effect on the dissemination of infectious bacteria throughout the world [bib_ref] Surveillance for travelrelated disease-GeoSentinel surveillance system, Harvey [/bib_ref]. This is particularly clear when the traveler is a sick person. For instance, Belgium received a sick patient travelling from Greece, and subsequently an outbreak caused by A. baumannii took place [bib_ref] Outbreak of infection by carbapenem-resistant Acinetobacter baumannii producing the carbapenemase OXA-58 in..., Bogaerts [/bib_ref]. Another clear case is the occurrence of A. baumannii outbreaks after the arrival of military patients from Iraq [bib_ref] Acinetobacter baumannii infections among patients at military medical facilities treating injured US..., Control [/bib_ref]. However, patient transfer between hospitals is not the single cause of international dissemination. Medical tourism has also had a strong impact as a cause of antibiotic-resistant bacteria dissemination in Latin America and Europe [bib_ref] Postsurgical wound infections due to rapidly growing mycobacteria in Swiss medical tourists..., Maurer [/bib_ref].
## Surveillance networks: advantages and current limitations
A positive outcome of the establishment of international surveillance networks is that member countries have to 6 Journal of Pathogens provide information. This baseline information is useful to understand the development and dissemination of antibioticresistant strains. Furthermore, the main positive impact perceived is the standardization of methods across countries and an improved laboratory capacity, both of which enhance the quality of testing. This is important because surveillance information allows the establishment of specific strategies targeting each country and/or region. These strategies can be designed according to impact, and they can be followed by particular indicators that can be measured at the surveillance centers belonging to the network. Despite the progress achieved so far, bacterial resistance to antibiotics is still a major public health problem. Reports of carbapenem resistance continue to increase. Therefore, surveillance networks must act on a continuous basis and in a comprehensive way. The sustainability of the surveillance networks depends on the support of the participant governments, including the on-going funding of network activities as well as adequate public health policies. The current situation of bacterial resistance to antibiotics has alarmed the international medical community. In consequence, World Health Organization (WHO) has developed a new plan to address the challenge posed by antimicrobial resistance.
## Current challenges to global surveillance integration
Different countries face different challenges; here we illustrate that although continuous surveillance programs are ideal, contributions of individual hospitals and research groups can provide valuable information. Here available data on Mexico is described as an example to illustrate the value of these efforts. An early limitation in the case of Mexico (1990s) was the use of phenotypic methods; as a consequence, such information should at least be used with caution when integrated with more recent data [bib_ref] A hospital epidemic due to Achromobacter calcoaceticus, Lecocq [/bib_ref] [bib_ref] Programa de monitoreo bacteriológico y de regulación de uso de antibióticos. Experiencia..., Rodríguez-Badillo [/bib_ref]. A summary of more recent data regarding A. baumannii in this country is presented in [fig_ref] Table 1: Surveillance studies carried out in Mexico [/fig_ref]. These data include surveillance carried out by individual hospitals; for instance, in a provincial hospital, MDR A. baumannii incidence was 74% for 2010 [bib_ref] Prevalence of multidrug-resistant bacteria at a tertiary-care teaching hospital in Mexico: special..., Garza-González [/bib_ref] and it became the second most frequently isolated pathogen [bib_ref] Acinetobacter baumannii infections in a tertiary care hospital in mexico over the..., Morfín-Otero [/bib_ref]. Also, the mortality rate imputable to this organism was 14.5% [bib_ref] Genetic characterisation of drug resistance and clonal dynamics of Acinetobacter baumannii in..., Bocanegra-Ibarias [/bib_ref]. Two additional studies were carried out by SENTRY: the 2003 study was inconclusive due to insufficient sampling, and the 2012 study showed that, at a national level, 16.7% of the isolates were imipenemresistant [bib_ref] Resistance trends of Acinetobacter spp. in Latin America and characterization of international..., Tognim [/bib_ref] [bib_ref] Antimicrobial resistance among Gram-negative bacilli isolated from Latin America: results from SENTRY..., Gales [/bib_ref]. In this country, regional surveillance is usually reported by the Ministry of Health (Secretaría de Salud (SESA)) [bib_ref] Informe anual de la Red de Monitoreo/Vigilancia de la Resistencia a los..., Ops-Usaid [/bib_ref]. There are two additional programs, the epidemiological hospital network (RHOVE)and the National Institute for Epidemiologic Reference (Instituto Nacional de Referencia Epidemiológica (INDRE)) program, as part of the Latin American network for monitoring resistance to antibiotics (ReLAVRE). However, no recent information on A. baumannii was available from these sources [bib_ref] Informe anual de la Red de Monitoreo/Vigilancia de la Resistencia a los..., Ops-Usaid [/bib_ref]. Nevertheless, the scarce data provided by a few studies reveal the presence of antibiotic-resistant Acinetobacter in hospital settings in Mexico. Furthermore, a multicenter study carried out recently provides valuable information to be integrated into the global picture. This study reported the presence of carbapenem-resistant A. baumannii in various hospitals from different regions in this country. It found a predominance of the Iberian-American clonal complex CC636 and the international clonal complex CC92. It also reported a regional distribution of oxacilinases OXA-72 and OXA-239 in the North and South of Mexico, respectively. An important observation derived from this study is the presence of different STs constituting two clonal complexes among the hospitals included. It is important to note that this kind of distribution had been previously reported in the USA, in Brazil [bib_ref] Emergence of Acinetobacter baumannii international clone II in Brazil: reflection of a..., Martins [/bib_ref] , and more recently in Argentina [bib_ref] Widespread dispersion of the resistance element tet(B)::ISCR2 in XDR Acinetobacter baumannii isolates, Vilacoba [/bib_ref]. In addition, regional distribution of OXAs in different STs suggests a process of dissemination among hospitals in different regions. This distribution may have arisen as a consequence of increased patient mobility, highlighting the relevance for further efforts towards a more integral surveillance and application of preventive strategies. Data provided by this kind of studies can be integrated into the global picture, as they contribute valuable information for the international surveillance of antibiotic-resistant A. baumannii.
## The role of surveillance and reporting in preventing the consequences of global mobilization
Increased mobilization of people worldwide is an intrinsic component of globalization. The increase in medical tourism increases the probability of MDR bacteria exchange. A clear example is the case of Mexico, which is considered one of the top medical tourism destinations [bib_ref] Medical tourism: globalization of the healthcare marketplace, Horowitz [/bib_ref] [bib_ref] Medical tourismhealth care in the global economy, Horowitz [/bib_ref]. As a consequence of frequent patient exchange, other countries have reported bacterial infections in patients arriving from Mexico [bib_ref] Postsurgical wound infections due to rapidly growing mycobacteria in Swiss medical tourists..., Maurer [/bib_ref]. Therefore, a key element in preventing global dissemination is the establishment of consistent surveillance programs capable of constant monitoring at sites with the highest patient exchange indices [bib_ref] Surveillance for travelrelated disease-GeoSentinel surveillance system, Harvey [/bib_ref]. Since bacterial antibiotic resistance is a global challenge, it would be ideal to ensure the cooperation of international and national surveillance instances. Local data collection and prevention are relevant when possible, but they are insufficient to compose a picture of the national situation and its consequences at a global scale. Further efforts, including potentially compulsory surveillance and data gathering, are required in order to properly face this global challenge. Bacteria causing specific diseases should be closely monitored through the establishment of precise programs. Such has been the case with tuberculosis: multidisciplinary action groups, like the Directly Observed Treatment Short-Course (DOTS) program, have established appropriate programs to prevent the spread of resistance. This strategy includes five important components: government, case detection, standardized treatment observed by healthcare workers, drug supply, and a standardized recording and reporting system. The implementation of a public health policy at an international level increases the potential success of the intervention [bib_ref] The politics of 'branding' in policy transfer: the case of DOTS for..., Ogden [/bib_ref]. The restricted sale of antibiotics is a unique public policy directed to stop the emergence of bacterial resistance in the community. The implementation of this policy is a key step in the global control of antibiotic resistance. However, additional factors may play important roles in limiting the positive effects of this policy. These potential factors may include the following: corruption or lack of law enforcement, governance, lack of surveillance, and lack of containment of the spread of resistant bacteria [bib_ref] Antimicrobial resistance: the major contribution of poor governance and corruption to this..., Collignon [/bib_ref]. Multicenter collaboration among hospitals remains a relevant component of surveillance and its continuity should be supported. However, it is critical to start working towards the establishment of a national surveillance program as part of the Global Strategy for Containment of Antimicrobial Resistance in medical centers. This global strategy has two critical aims. The first one is to prevent resistant bacteria dissemination; the second one is to prevent the accelerated emergence of new forms of resistance in order to avoid further restriction of therapeutic options. The main mechanism to contain nosocomial outbreaks has been the use of antibiotics. However, when antibiotics are rendered ineffective due to bacterial resistance, the possibility of an efficient containment becomes compromised. In order to propose, implement, and evaluate new specific containment interventions it is essential to gather specific data on MDR bacteria such as A. baumannii and make them available.
[fig] Figure 1: Top diagram shows dates of introduction of antimicrobials; relevant examples are indicated below colored lines for some groups. Dashed lines indicate antibiotics without group. Insert graph shows the date of introduction of antimicrobials, date of first reports of antimicrobial resistance in A. baumannii, and the emergence of multidrug-resistant (MDR), extensively drug-resistant (XDR), and pandrugresistant (PDR) bacterial strains. Dashed lines indicate interval between antibiotic introduction and first report of resistance in A. baumannii. [/fig]
[fig] Figure 2: Brief history of the incorporation of Acinetobacter baumannii as one of the successful multidrug-resistant nosocomial pathogens. This has led to the strategic alliance of different countries and continents in monitoring resistant bacteria, and it has resulted in the formulation of a new action plan against bacterial resistance in 2015. [/fig]
[table] Table 1: Surveillance studies carried out in Mexico. Predominance of clonal complexes CC636 and CC92. New OXA-469 allele found and regional distribution of OXA-72 and OXA-239 determined. Submitted a Included all the states in Mexico. +, positive; −, negative; SLP, San Luis Potosí. [/table]
|
Phosphorylated ERK is a potential predictor of sensitivity to sorafenib when treating hepatocellular carcinoma: evidence from an in vitro study
Background: Sorafenib is the first agent that has demonstrated an improved overall survival benefit in advanced hepatocellular carcinoma (HCC), setting a new standard for first-line treatment. However, no one has yet been able to predict sensitivity to sorafenib. Pre-treatment pERK level has been shown to be associated with favorable response to such therapy in a phase II clinical study, indicating that pERK may be a potential biomarker for treatment of HCC with sorafenib.Methods:The effects of sorafenib and 5-fluorouracil (5-FU) on cell proliferation were evaluated by cell viability assays in four HCC cell lines (SMMC-7721, MHCC97-L, MHCC97-H and HCCLM6) with different metastatic potential and basal pERK expression levels. Expression levels of pERK were determined by immunocytochemical quantification together with western blot analysis, and pERK density values were also calculated. Correlation analyses were then carried out between the IC 50 values of drugs and pERK density values. After basal ERK phosphorylation was down-regulated with U0126 in MHCC97-H cells, cellular responsiveness to sorafenib was assessed by cell viability assay.Results: Basal pERK levels increased stepwise in cell lines in accordance with their metastatic potential. Sorafenib inhibited ERK phosphorylation in a dose-dependent manner in all four cell lines at a concentration between 5 and 20 μM, but the degree of inhibition was significantly different according to their basal pERK expression level (P < 0.0001). In contrast, no significant change was observed after 5-FU treatment. Correlation analyses between the IC 50 values and pERK densities revealed that the effects of sorafenib on cell proliferation were significantly correlated with basal pERK levels (Spearman r = -0.8671, P = 0.0003). Resistance to 5-FU was also significantly associated with basal pERK expression in these HCC cell lines (Spearman r = 0.7832, P = 0.0026). After the basal ERK phosphorylation level in MHCC97-H cells was reduced with U0126, they were significantly less sensitive to sorafenib-mediated growth inhibition, with an IC 50 of 17.31 ± 1.62 μM versus 10.81 ± 1.24 μM (P = 0.0281).
# Background
Hepatocellular carcinoma (HCC) is the sixth most common malignancy worldwide and the third most common cause of death from cancer, accounting for more than 626,000 new cases and 598,000 deaths per year. Of all these cases, more than half are in China alone [bib_ref] Global cancer statistics, Parkin [/bib_ref]. The disease is diagnosed at early stages in 30 to 40% of all patients and is amenable to potentially curative treatments, such as surgical therapies (resection and liver transplantation) and locoregional procedures (radiofrequency ablation). Five-year survival rates of up to 60 to 70% can be achieved in well-selected patients. However, disease diagnosed at an advanced stage or with progression after locoregional therapy has a dismal prognosis, owing to the underlying liver disease and lack of effective treatment options [bib_ref] Management of hepatocellular carcinoma, Bruix [/bib_ref]. No systemic therapy with traditional chemotherapy drugs has improved survival in patients with advanced hepatocellular carcinoma [bib_ref] Systematic review of randomized trials for unresectable hepatocellular carcinoma: Chemoembolization improves survival, Llovet [/bib_ref].
Sorafenib (Nexavar, Bayer HealthCare Pharmaceuticals) is an oral multikinase inhibitor that inhibits the serine-threonine kinases Raf-1 and B-Raf, the receptor tyrosine kinase activity of vascular endothelial growth factor (VEGF) receptors 1, 2, and 3, and platelet-derived growth factor receptor β [bib_ref] BAY 43-9006 exhibits broad spectrum oral antitumor activity and targets the RAF/MEK/ERK..., Wilhelm [/bib_ref]. It blocks tumor cell proliferation and tumor angiogenesis, and increases the rate of apoptosis in a wide range of tumor models by targeting the Raf/ mitogen-activated protein kinase kinase/extracellular signal-regulated kinase (RAF/MEK/ERK) and VEGF signaling pathways [bib_ref] Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell..., Liu [/bib_ref]. The results of a phase III, randomized, placebo-controlled trial, the Sorafenib HCC Assessment Randomized Protocol (SHARP) trial, were recently presented [bib_ref] SHARP Investigators Study Group: Sorafenib in advanced hepatocellular carcinoma, Llovet [/bib_ref]. In this trial, sorafenib demonstrated improved overall survival and time to tumor progression in patients with advanced HCC. This landmark study represents the first agent that has demonstrated an improved overall survival benefit in this disease and sets a new standard for the firstline treatment of advanced HCC that has been approved by the US Food and Drug Administration (FDA).
However, no one has yet predicted sensitivity to sorafenib in the treatment of HCC. It is well known that phosphorylated ERK (pERK) is a key downstream component of the RAF/MEK/ERK signaling pathway. It can be translocated to the nucleus after phosphorylation, where it leads to changes in gene expression by phosphorylating and regulating various transcription factors, such as Ets family transcription factors (for example, Elk-1) [bib_ref] Targeting the Raf-MEK-ERK mitogen-activated protein kinase cascade for the treatment of cancer, Roberts [/bib_ref]. In a phase II study in 137 patients with advanced, inoperable HCC, of which 33 had their pre-treatment pERK levels evaluated, pre-treatment tumor pERK levels were correlated with the time to tumor progression. Patients whose tumors expressed higher baseline pERK levels had a longer time to tumor progression following treatment with sorafenib [bib_ref] Phase II study of sorafenib in patients with advanced hepatocellular carcinoma, Abou-Alfa [/bib_ref]. These data suggest that tumors containing higher levels of pERK are more sensitive, or responsive, to sorafenib, indicating that pERK may be a useful biomarker in treating HCC with sorafenib. Whether this marker will prove to be predictive of response needs to be validated in future studies.
To investigate the relationship between the effects of sorafenib on cell proliferation and basal pERK levels in HCC cell lines, here we evaluate the effects of sorafenib on four HCC tumor cell lines (SMMC-7721, MHCC97-L, MHCC97-H and HCCLM6) with different metastatic potentials and baseline pERK expression levels. A series of human HCC cell lines with similar genetic backgrounds yet dramatic differences in spontaneous metastatic behaviors, which had been established at the authors' institute [bib_ref] Stepwise metastatic human hepatocellular carcinoma cell model system with multiple metastatic potentials..., Li [/bib_ref] [bib_ref] New human hepatocellular carcinoma (HCC) cell line with highly metastatic potential (MHCC97)..., Tian [/bib_ref] , provided a unique platform for this research. Among these cell lines, SMMC-7721 is low-invasive and non-metastatic. MHCC97-L and MHCC97-H are two different metastatic HCC cell clones isolated from the same parent cell line MHCC97, which was derived from a nude mouse model of human HCC metastasis (LCI-D20). The LCI-D20 model was developed by orthotopic inoculation of an intact tumour tissue of an intrahepatic disseminated lesion from a 39-year-old Chinese male patient with HCC (Zhongshan Hospital, Fudan University, Shanghai, China) in whose serum abnormal alpha-fetoprotein and HBsAg (hepatitis B surface antigen) were found. Spontaneous pulmonary metastasis occurred in 40% and 100% of recipient nude mice after orthotopic transplantation of MHCC97-L and MHCC97-H, respectively. HCCLM6 was established from MHCC97-H by six rounds of in vivo metastasis selection and produced further multiple extensive metastases through both blood vessels and lymphatic channels. Such characteristics make these cell lines valuable for comparative study.
# Materials and methods
## Drug preparations
[formula] Sorafenib tosylate (Nexavar, [N-(3-trifluoromethyl-4- chlorophenyl)-N-(4-(2-methylcarbamoylyridin-4-yl)oxy- phenyl)urea]) [/formula]
## Immunocytochemical staining and quantification
Cells were plated in six-well plates with cover slips at 4 × 10 5 per well. On the following day, cells were treated with compounds indicated in the experiment. Briefly, cells were exposed to 5, 10, or 20 μM sorafenib for 24 hours. Cells were exposed to 20 μM U0126 for 6 hours. DMSO was added to cultures at 0.1% (v/v) as a solvent control. Cells were treated with 10, 20, or 50 mg/l 5-fluorouracil (5-FU) for 48 hours. Cell culture medium without 5-FU was used as a control. After being fixed in acetone and blocked serially with IHC Biotin Block kit, 3% H 2 O 2 , and 10% normal goat serum, sections were incubated with the mouse monoclonal antibody to ERK1 + ERK2 (Clone MAPK-YT, Abcam, Cambridge, MA, USA) at 1:100 dilution overnight at 4°C. The UltraSensitive™ S-P stain system (Maixin Biotechnology Development Co., Ltd, Fuzhou, China) was applied according to the manufacturer's instructions. Sections were then developed in diaminobenzidine solution and counterstained with Mayer's hematoxylin. Negative controls were performed by omitting the primary antibodies.
Sections were observed at 200× magnification in a computerized image system composed of a Leica CCD camera DFC420 connected to a Leica DM IRE2 microscope (Leica Microsystems Imaging Solutions Ltd, Cambridge, United Kingdom) and images were captured by the Leica QWin Plus version 3 software under the same conditions. The same protein quantification method was used for pERK quantification with Image-Pro Plus version 6.2 software (Media Cybernetics Inc., Bethesda, MD, USA) as Sun's group reported [bib_ref] High expression of macrophage colony-stimulating factor in peritumoral liver tissue is associated..., Zhu [/bib_ref]. The pERK density in each field was calculated as [Integrated optical density of pERK-positive objects/(Total field area -Blank area)]. The mean value of pERK density in each group was calculated on six random field samples from three independent experiments with replicates per experiment. The expression rate of pERK was calculated from pERK density and the expression rate in the control group of each cell line was set as the 100% baseline. Data within each group were analyzed statistically with one-way ANOVA and differences between cell lines of sorafenib's pERK inhibition were analyzed by two-way ANOVA, both of which were followed by Bonferroni's multiple comparison test with SPSS 13.0 for Windows (SPSS Inc. Chicago, IL, USA). P < 0.05 was considered significant.
# Immunoblot analysis
Cells were plated at 6 × 10 5 cells per well in six-well plates. On the following day, cells were treated with the same methods as described above. After treatment, cells were washed with cold phosphate-buffered saline and lysed using RIPA lysis buffer containing 1 mM phenylmethylsulfonyl fluoride (PMSF). Twenty micrograms of protein, which was determined using a bicinchoninic acid protein assay, from control and treated cell lysates was loaded on 5% and 12% SDS-PAGE gels, electrophoresed at a constant voltage of 70 V for 2 hours, and transferred onto PVDF membranes (0.45 μm) at a constant voltage of 80 V for 2.5 hours. Blots were probed with a 1:1,000 dilution of mouse monoclonal to ERK1 + ERK2 antibody (Clone MAPK-YT, Abcam, Cambridge, MA, USA), a 1:3,000 dilution of anti-human β-actin monoclonal antibody, then horseradish peroxidase-conjugated secondary antibody (1:500) and detected by enhanced chemiluminescence reagent (ECL kit, Pierce, Rockford, IL, USA). Unless otherwise indicated, immunoblot reagents were purchased from Beyotime Institute of Biotechnology (Shanghai, China).
## Cell viability assay
Cells were plated at 5,000 cells per well in 96-well microtiter plates and incubated overnight at 37°C in a humidified incubator containing 5% CO 2 . On the following day, compounds were added to the wells indicated in the experiment. Cells were exposed to sorafenib for 24 hours at concentrations of 0.01, 0.1, 1, 2, 4, 5, 10, 15, 20, 25 or 30 μM, and to U0126 for 6 hours at concentrations of 1, 5, 10, 20, 50 or 100 μM. In the sequential combination experiment, cells were pretreated with 20 μM U0126 for 6 hours and then exposed to sorafenib for a further 24 hours. DMSO was added to cultures at 0.1% (v/v) as a solvent control. Cells were treated with 5-FU for 48 hours at concentrations of 0.01, 0.1, 1, 5, 10, 20, 50, 100, 200, 500 or 1,000 mg/l. Cell culture medium without 5-FU was used as a control. Cell viability was determined using the Cell Counting Kit-8 (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer's instructions [bib_ref] Transcriptional activation of the thyroglobulin promoter directing suicide gene expression by thyroid..., Shimura [/bib_ref]. IC 50 values were calculated by nonlinear regression analysis using GraphPad Prism version 5.0 software (GraphPad Software, Inc., San Diego, CA, USA) [bib_ref] Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell..., Liu [/bib_ref] , according to the results of at least three independent experiments with four replicates of each cell line per experiment. Differences in cellular responsiveness to drugs were analyzed statistically with two-way ANOVA with SPSS 13.0 for Windows (SPSS Inc.). Spearman's rank correlation method was used for correlation analyses between pERK density values and drugs' IC 50 values of three independent experiments for four cell lines with four replicates each (SPSS 13.0 for Windows, SPSS Inc.). P < 0.05 was considered significant.
# Results
## Basal perk levels in hcc cell lines increase stepwise with their metastatic potential
Basal pERK levels in four HCC cell lines were measured by immunocytochemistry and image quantification. Immunocytochemical analysis showed that pERK proteins were found in both the nuclei and cytoplasm of tumor cells. However, pERK in cell lines with higher metastatic potential seemed more inclined to be located in the nucleus, with stronger staining intensity [fig_ref] Figure 1: Basal pERK levels in hepatocellular carcinoma [/fig_ref]. The results of image quantification confirmed that baseline pERK was differentially expressed in these HCC cell lines (P < 0.0001, n = 6) and seemed to be correlated with their metastatic potential. The pERK density in SMMC-7721, MHCC97-L, MHCC97-H and HCCLM6 cells was 0.042 ± 0.006, 0.081 ± 0.007, 0.329 ± 0.037 and 0.463 ± 0.084, respectively. In metastatic MHCC97-H and HCCLM6 cells, pERK levels were significantly higher than in nonmetastatic SMMC-7721 cells (P < 0.0001, n = 6). Even among the three metastatic cell lines, pERK levels were differentially expressed and increased stepwise with their metastatic potential (P < 0.0001, n = 6; [fig_ref] Figure 1: Basal pERK levels in hepatocellular carcinoma [/fig_ref].
Baseline ERK phosphorylation levels in these cancer cells were also examined by western blot analysis. Consistent with immunocytochemical analysis, the results demonstrated that cancer cells with more invasive potential such as HCCLM6 and MHCC97-H cells expressed higher levels of pERK when compared to the relatively less invasive MHCC97-L or SMMC-7721 cells [fig_ref] Figure 1: Basal pERK levels in hepatocellular carcinoma [/fig_ref].
## Effects of sorafenib on erk phosphorylation inhibition are significantly associated with basal perk levels in hcc cell lines
The pERK protein is best known as a key downstream component of the RAF/MEK/ERK pathway. Changes in the levels of ERK phosphorylation were determined by immunocytochemical analysis in order to evaluate the effects of sorafenib on this pathway. In our study, sorafenib could inhibit ERK phosphorylation in all four HCC cell lines dose-dependently at a concentration between 5 and 20 μM . After exposure to 5, 10 or 20 μM sorafenib for 24 hours, the expression rate of pERK in SMMC-7721 cells fell gradually to 81.88 ± 7.65%, 71.63 ± 10.80% and 17.47 ± 1.34%, respectively, and in HCCLM6 cells to 78.06 ± 4.66%, 28.12 ± 1.36% and 3.99 ± 0.19%, respectively . The expression rates in both cell lines were significantly reduced when compared to each DMSO control group (P = 0.0043, n = 6, and P < 0.0001, n = 6, respectively). However, further statistical analyses revealed the significant difference in the degree of the sorafenib effects in these HCC cell lines. Interestingly, the sorafenib pERK inhibition effect in SMMC-7721 cells with lower initial levels of pERK was significantly weaker when compared to the other three HCC cell lines with relatively higher basal pERK levels (P < 0.0001, n = 6; , and it should be noted that this difference was mainly at 10 μM sorafenib. No significant difference was found in MHCC97-L, MHCC97-H and HCCLM6 cells.
On the contrary, no significant change was observed after 5-FU treatment in MHCC97-H cells . The pERK expression rate was 102.3 ± 7.88%, 110.8 ± 6.60%, and 101.1 ± 5.12%, respectively, after exposure to 10, 20 or 50 mg/l 5-FU for 48 hours, with no statistical difference with the control group (P > 0.05, n = 6; . Western blot analysis confirmed the same results above .
## Effects of sorafenib on cell proliferation are significantly correlated with basal perk levels in hcc cell lines
The effects of sorafenib on cell proliferation were measured by the CCK-8 cell viability assay. According to our results, sorafenib inhibited proliferation of all four HCC cell lines in a dose-dependent manner as described in previous research [bib_ref] Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell..., Liu [/bib_ref] , with an IC 50 of 20.85 ± 2.81 μM, 10.38 ± 1.52 μM, 10.70 ± 2.35 μM and 9.11 ± 2.44 μM in SMMC-7721, MHCC97-L, MHCC97-H and HCCLM6 cells, respectively. As expected, SMMC-7721 cells, which contain the lowest levels of pERK, were significantly less sensitive to sorafenib-mediated growth inhibition than the other three HCC cell lines with higher basal pERK levels (P = 0.0018, n = 4; [fig_ref] Figure 4: Effects of sorafenib and resistance to 5-fluorouracil [/fig_ref]. Meanwhile, a significant negative correlation (Spearman r = -0.8671, P = 0.0003, n = 12) was observed between the IC 50 values of sorafenib in these HCC cell lines and their pERK density values [fig_ref] Figure 4: Effects of sorafenib and resistance to 5-fluorouracil [/fig_ref] , indicating that the effects of sorafenib on cell proliferation were significantly correlated with basal pERK levels in these HCC cell lines.
Opposite results were observed with treatment with the traditional chemotherapy drug 5-FU. 5-FU inhibited HCC cell proliferation with an IC 50 of 4.24 ± 0.87 mg/l, 79.71 ± 24.49 mg/l, 41.21 ± 21.55 mg/l and 187.45 ± 78.05 mg/ l in SMMC-7721, MHCC97-L, MHCC97-H and HCCLM6 cells, respectively, with significant statistical differences (P < 0.0001, n = 4). The SMMC-7721 cells, with lower pERK expression, demonstrated a higher sensitivity to 5-FU. However, MHCC97-L, MHCC97-H, and HCCLM6 cells, with higher pERK expression, exhibited more resistance to this drug. The ultimate inhibition rate before reaching a plateau in these three cell lines was about 35%, 40%, and 45%, respectively, each compared to its control group [fig_ref] Figure 4: Effects of sorafenib and resistance to 5-fluorouracil [/fig_ref]. Furthermore, a significant correlation (Spearman r = 0.7832, P = 0.0026, n = 12) was observed between Basal pERK levels in hepatocellular carcinoma (HCC) cell lines increased stepwise in accordance with their metastatic potential The degree of sorafenib's inhibition of ERK phosphorylation was significantly associated with basal pERK levels in hepatocellular carcinoma (HCC) cell lines The degree of sorafenib's inhibition of ERK phosphorylation was significantly associated with basal pERK levels in hepatocellular carcinoma (HCC) cell lines. (A) Immunocytochemical staining of pERK (200×) after treatment with sorafenib for 24 hours in four HCC cell lines. (B) pERK expression rate in HCC cell lines after sorafenib treatment. pERK levels were reduced after sorafenib treatment in all four cell lines, but the inhibition effect in SMMC-7721 cells, with lower levels of pERK, was significantly weaker when compared to the other three HCC cell lines with relatively higher basal pERK levels (two-way ANOVA, P < 0.0001, n = 6). The expression rate of pERK was calculated from pERK density determined as described in Materials and methods and the rate in each control group was set as the 100% baseline. Columns represent means of six samples in each group; bars indicate standard error. the IC 50 values of 5-FU and pERK density values [fig_ref] Figure 4: Effects of sorafenib and resistance to 5-fluorouracil [/fig_ref] , indicating that the resistance to 5-FU was significantly associated with basal pERK expression in these HCC cell lines.
## 5-fluorouracil (5-fu) hardly inhibited erk phosphorylation in mhcc97-h cells
## Effects of mek/erk pathway inhibition and perk reduction on sensitivity to sorafenib
To more directly determine the relationship between pERK expression and sensitivity to sorafenib, we inhibited the MEK/ERK pathway and reduced basal pERK expres-sion in MHCC97-H cells via U0126, a selective inhibitor of MEK 1 and MEK 2 [bib_ref] Potentiation of the antitumor effects of both selective cyclooxygenase-1 and cyclooxygenase-2 inhibitors..., Cusimano [/bib_ref] , and then compared cellular responsiveness to sorafenib to that of untreated cells. To avoid possible direct growth inhibition by U0126, exposure of cells to this drug was limited to 6 hours according to our preliminary experiments. Quantification of cellular pERK levels by immunocytochemical analysis indicated that constitutive ERK phosphorylation was strongly reduced in MHCC97-H cells after treatment with 20 μM U0126 for 6 hours, relative to the level observed in the Effects of sorafenib and resistance to 5-fluorouracil (5-FU) were significantly correlated with basal pERK levels in hepatocellular carcinoma (HCC) cell lines value, at which 50% of cell growth is inhibited compared with control, was calculated by nonlinear regression analysis using GraphPad Prism version 5.0 software. pERK density was quantified using Image-Pro Plus version 6.2 software as described in Materials and methods. Spearman's rank correlation method was used for correlation analysis between pERK density values and drugs' IC 50 values of three independent experiments for four cell lines with four replicates each. P < 0.05 was considered significant.
untreated cells (60% inhibition; [fig_ref] Figure 3 5 -: Fluorouracil [/fig_ref] , which induced almost no detectable systemic toxicity on cell proliferation [fig_ref] Figure 3 5 -: Fluorouracil [/fig_ref].
In the following experiments, we compared sorafenib responsiveness of MHCC97-H cells pretreated with 20 μM U0126 for 6 hours to an untreated control. Cell viability assay revealed that the pretreated cells were significantly less sensitive to sorafenib-mediated growth inhibition, with an IC 50 of 17.31 ± 1.62 μM versus 10.81 ± 1.24 μM (P = 0.0281, n = 6; [fig_ref] Figure 3 5 -: Fluorouracil [/fig_ref]. These results confirmed that the RAF/MEK/ERK signaling pathway was essential for sorafenib-mediated growth inhibition, and that the sensitivity to sorafenib was directly related to the activation of this pathway and basal pERK expression in MHCC97-H cells.
# Discussion
It is well known that the RAF/MEK/ERK cascade is a key signaling pathway involved in the regulation of normal mammalian cell proliferation, survival and differentiation. It couples signals from cell surface receptors to transcription factors and regulates gene expression though a phosphorylation cascade. Raf serine/threonine kinases phosphorylate and activate the MEK1/2 dual-specificity protein kinases, which then phosphorylate and activate ERK1/2. Activated ERK is a downstream component of an evolutionarily conserved signaling module that can be translocated to the nucleus, where it phosphorylates and regulates various transcription factors, ultimately leading to changes in gene expression. Additionally, Ras family small GTPases are key upstream activators of the RAF/ MEK/ERK pathway, which are often activated by upstream molecules such as receptor tyrosine kinases (for example, epidermal growth factor receptor, VEGF receptor and transforming growth factor-α receptors) [bib_ref] Targeting the Raf-MEK-ERK mitogen-activated protein kinase cascade for the treatment of cancer, Roberts [/bib_ref]. Mutation or over-activation of related components in the RAF/MEK/ ERK cascade would lead to acceleration of cell proliferation and extension of survival, thus contributing to human oncogenesis [bib_ref] Roles of the RAF/MEK/ERK and PI3K/PTEN/AKT pathways in malignant transformation and drug..., Mccubrey [/bib_ref]. This pathway has been implicated in the molecular pathogenesis of HCC. First of all, as an upstream activator of this pathway, the Ras gene is mutationally activated in 30% of HCCs [bib_ref] Targeting RAS signalling pathways in cancer therapy, Downward [/bib_ref]. Second, Raf kinase over-expression occurs in most HCCs. For example, in a study on HCC tissue specimens, the Raf-1 gene was up-regulated in 50% of 22 HCC specimens and activated Raf-1 protein was overexpressed in 100% of 30 HCC specimens [bib_ref] Over-expression of c-raf-1 proto-oncogene in liver cirrhosis and hepatocellular carcinoma, Hwang [/bib_ref]. Third, a variety of upstream growth factors, such as epidermal growth factor, VEGF, platelet-derived growth factor-β and transforming growth factor-α, which are generally overexpressed in HCC, can activate this pathway through binding their receptor tyrosine kinases [bib_ref] Targeting the Raf-MEK-ERK mitogen-activated protein kinase cascade for the treatment of cancer, Roberts [/bib_ref] [bib_ref] Angiogenesis and hepatocellular carcinoma, Semela [/bib_ref].
The pERK protein is a key downstream component of the MEK/ERK cascade. In this study, basal levels of pERK were determined by immunocytochemical analysis and western blot analysis in order to evaluate the activation of the RAF/MEK/ERK pathway in four types of HCC cell lines with different metastatic potential. The results revealed that basal pERK levels increased stepwise in cell lines in accordance with their metastatic potential, indicating that the RAF/MEK/ERK pathway may be involved in tumor invasion and metastasis in HCC, consistent with results of previous studies [bib_ref] Role of MAP kinase in tumor progression and invasion, Reddy [/bib_ref].
Sorafenib is a multikinase inhibitor that inhibits the Raf serine-threonine kinases and blocks the RAF/MEK/ERK signaling pathway. Changes in the levels of pERK protein were determined by immunocytochemical analysis in these HCC tumor cells. Sorafenib inhibited ERK phosphorylation in a dose-dependent manner between 5 and 20 μM. However, further analyses revealed that the degree of inhibition in these HCC cell lines was significantly different according to their basal pERK expression levels. We found that the sorafenib pERK inhibition effect in SMMC-7721 cells, with lower pERK levels, was significantly weaker than the other three HCC cell lines with relatively higher basal pERK levels. On the contrary, no significant change in pERK phosphorylation was observed after 5-FU treatment.
It is possible that the antitumor activity of sorafenib might be due to its ability to inhibit angiogenesis-related tyrosine kinases as well as other RAF/MEK/ERK-independent mechanisms. For example, Raf-1 has been proposed to induce the phosphorylation of proteins that control apoptosis independently of MEK and ERK [bib_ref] Roles of the RAF/MEK/ERK and PI3K/PTEN/AKT pathways in malignant transformation and drug..., Mccubrey [/bib_ref]. Additionally, the results of clinical trial analyses of sorafenib in renal cell carcinoma and melanoma have not provided sufficient information to conclude that the clinical value is associated with inhibition of the RAF/MEK/ERK signaling pathway [bib_ref] Targeting the Raf-MEK-ERK mitogen-activated protein kinase cascade for the treatment of cancer, Roberts [/bib_ref].
However, positive results were observed in this study. In cell viability assays, sorafenib inhibited proliferation of all HCC cell lines with different basal pERK levels in a dosedependent manner. Furthermore, the effects of sorafenib were significantly correlated with basal pERK levels in these HCC cell lines by correlation analysis between the IC 50 values of sorafenib and their pERK density values, indicating that sorafenib sensitivity could have direct links with the activation of the RAF/MEK/ERK signaling pathway and basal pERK levels in HCC tumor cells.
To more directly determine the relationship between pERK expression and sensitivity to sorafenib, we used U0126, a selective inhibitor of MEK 1/2, to inhibit the Effects of reducing basal pERK levels on sorafenib sensitivity μM. (C) pERK expression was reduced after U0126 treatment in MHCC97-H cells. The expression rate of pERK was calculated from pERK density determined as described in Materials and methods and the rate in the control group was set as the 100% baseline. Columns represent means from six samples in each group; bars indicate standard deviation; *, P < 0.05, Student's t-test was used when compared with solvent control. (D) Effects of U0126 treatment for 6 hours on cell proliferation in MHCC97-H cells. (E) Effects of sorafenib individually and in sequential combination with U0126 on cell proliferation in MHCC97-H cells. In the sequential combination experiments, cells were pre-treated with 20 μM U0126 for 6 hours and then exposed to sorafenib for a further 24 hours. Each value represents the average of six independent determinations with four replicates per experiment; bars indicate standard error.
## Mek/erk pathway and reduce basal perk expression in mhcc97-h cells without influencing cell proliferation.
We then assessed cellular responsiveness to sorafenib after pERK down-regulation. The observations showed that the pre-treated cells expressed much lower levels of pERK and became significantly less sensitive to sorafenib-mediated growth inhibition. These observations are perfectly consistent with our hypothesis that the RAF/MEK/ERK signaling pathway is essential for sorafenib-mediated growth inhibition and that sensitivity to sorafenib is directly related to the activation of this pathway and basal pERK expression. These results also confirm the findings of Ghassan K Abou-Alfa's group in a phase II clinical trial on treating advanced HCC patients with sorafenib that found that patients with tumors containing higher levels of pERK were more sensitive, or responsive, to sorafenib, supporting the notion that pERK may be a useful biomarker in treating HCC with sorafenib [bib_ref] Phase II study of sorafenib in patients with advanced hepatocellular carcinoma, Abou-Alfa [/bib_ref]. In our opinion, this was probably because blocking of the RAF/MEK/ERK signaling pathway plays a central role in sorafenib antitumor activity in HCC cells. As a key downstream component of this pathway, pERK levels could reflect the constitutive activity status of this signaling pathway, as well as the degree of inhibition of this pathway by sorafenib. Our study provides further in vitro evidence that pERK could be a useful biomarker predictive of a response to sorafenib in HCC tumor cells.
Resistance to the traditional chemotherapy drug 5-FU was closely associated with basal pERK expression in these HCC cell lines. Thus, the RAF/MEK/ERK pathway may be involved in the development of drug resistance to traditional chemotherapy in HCC, as reported in previous studies in other types of cancer. For instance, in breast cancer cells, activated Raf conferred resistance to the chemotherapeutic drugs doxorubicin and paclitaxel by inducing the expression of the drug pump Mdr-1 and the Bcl-2 antiapoptotic protein [bib_ref] Roles of the RAF/MEK/ERK and PI3K/PTEN/AKT pathways in malignant transformation and drug..., Mccubrey [/bib_ref]. The results reported here would provide clues for further studies on reducing drug resistance by blocking the RAF/MEK/ERK signaling pathway and rationally combining sorafenib with other traditional cytotoxic agents to further improve efficacy.
# Conclusion
These experiments demonstrate that the RAF/MEK/ERK pathway might be involved in drug resistance to traditional chemotherapy in HCC cell lines. More importantly, our study provides further in vitro evidence that pERK could be a useful biomarker predictive of sensitivity to sorafenib in HCC tumor cells.
## Abbreviations
DMEM: Dulbecco's modified Eagle's medium; DMSO: dimethyl sulfoxide; ERK: extracellular signal-regulated kinase; 5-FU: 5-fluorouracil; HCC: hepatocellular carci-noma; MEK: mitogen-activated protein kinase kinase; pERK: phosphorylated ERK; VEGF: vascular endothelial growth factor.
[fig] Figure 1: Basal pERK levels in hepatocellular carcinoma (HCC) cell lines increased stepwise in accordance with their metastatic potential. (A-E) Immunocytochemical staining of pERK (200×). (A) Negative control. (B) SMMC-7721. (C) MHCC97-L. (D) MHCC97-H. (E) HCCLM6. (F) Quantification of basal pERK levels in different HCC cell lines. pERK density was quantified using Image-Pro Plus version 6.2 software as described in Materials and methods. Columns represent means of six samples in each group; bars indicate standard deviation; *, P < 0.05, when compared to the negative control. (G) Western blot analysis of pERK protein in different HCC cell lines. [/fig]
[fig] Figure 3 5 -: Fluorouracil (5-FU) hardly inhibited ERK phosphorylation in MHCC97-H cells. (A-D) Immunocytochemical staining of pERK (200×) after treatment with 5-FU for 48 hours in MHCC97-H cells. (A) Culture medium without 5-FU was used as control. (B) 5-FU 10 mg/l. (C) 5-FU 20 mg/l. (D) 5-FU 50 mg/l. (E) pERK expression was not reduced after 5-FU treatment. The expression rate of pERK was calculated from pERK density determined as described in Materials and methods and the rate in the control group was set as the 100% baseline. Columns represenr means of six samples in each group; bars indicate standard deviation; *, P < 0.05, when compared to the control. (F) Western blot analysis of pERK protein in MHCC97-H cells after treatment with 5-FU for 48 hours. [/fig]
[fig] Figure 4: Effects of sorafenib and resistance to 5-fluorouracil (5-FU) were significantly correlated with basal pERK levels in hepatocellular carcinoma (HCC) cell lines. (A) Sorafenib inhibited cell proliferation in different HCC cell lines. (B) 5-FU inhibited cell proliferation in different HCC cell lines. Each value represents the average of four independent determinations with four replicates per experiment. Bars indicate standard error. (C) Correlation analysis between the IC 50 values of sorafenib and pERK density values. (D) Correlation analysis between the IC 50 values of 5-FU and pERK density values. The IC 50 [/fig]
[fig] Figure 5: Effects of reducing basal pERK levels on sorafenib sensitivity. (A, B) Immunocytochemical staining of pERK (200×) in MHCC97-H cells after treatment with 20 μM U0126 for 6 hours. (A) DMSO (0.1%) was used as solvent control. (B) U0126 20 [/fig]
|
Impact of Smoking and Chewing Tobacco on Arsenic-Induced Skin Lesions
[bib_ref] Groundwater arsenic contamination and its health effects in the Ganga-Meghna-Brahmaputra plain, Chakraborti [/bib_ref] [bib_ref] Association of arsenic exposure during pregnancy with fetal loss and infant death:..., Rahman [/bib_ref] [bib_ref] Arsenic and cardiovascular disease, States [/bib_ref] [bib_ref] Arsenic carcinogenesis in the skin, Yu [/bib_ref] [bib_ref] Modification of risk of arsenic-induced skin lesions by sunlight exposure, smoking, and..., Chen [/bib_ref] [bib_ref] Diagnosis of arsenicosis, Saha [/bib_ref] [bib_ref] Mechanisms of arsenic biotransformation, Vahter [/bib_ref] [bib_ref] Arsenic methylation and skin cancer risk in southwestern Taiwan, Chen [/bib_ref] [bib_ref] Arsenic methylation and bladder cancer risk in Taiwan, Chen [/bib_ref] [bib_ref] Biomarkers of exposure, effect, and susceptibility of arsenic-induced health hazards in Taiwan, Chen [/bib_ref] [bib_ref] Interaction between environmental tobacco smoke and arsenic methylation ability on the risk..., Chen [/bib_ref] [bib_ref] Urinary 8-hydroxydeoxyguanosine and urothelial carcinoma risk in low arsenic exposure area, Chung [/bib_ref] [bib_ref] Serum beta-carotene level, arsenic methylation capability, and incidence of skin cancer, Hsueh [/bib_ref] [bib_ref] Arsenic exposure, urinary arsenic speciation, and the incidence of urothelial carcinoma: a..., Huang [/bib_ref] [bib_ref] Urinary arsenic specia-Urinary arsenic speciation and its correlation with 8-OHdG in Chinese..., Li [/bib_ref] [bib_ref] Association between the clastogenic effect in peripheral lymphocytes and human exposure to..., Mäki-Paakkanen [/bib_ref] [bib_ref] Urinary arsenic profile affects the risk of urothelial carcinoma even at low..., Pu [/bib_ref] [bib_ref] Arsenic methylation and bladder cancer risk in case-control studies in Argentina and..., Steinmaus [/bib_ref] [bib_ref] Association of oxidative stress with arsenic methylation in chronic arsenicexposed children and..., Xu [/bib_ref] [bib_ref] Arsenic methylation capacity and skin cancer, Yu [/bib_ref] [bib_ref] Arsenic metabolism, genetic susceptibility, and risk of premalignant skin lesions in Bangladesh, Ahsan [/bib_ref] [bib_ref] The risk of arsenic induced skin lesions in Bangladeshi men and women..., Lindberg [/bib_ref] [bib_ref] Lung cancer and arsenic exposure in rural Bangladesh, Mostafa [/bib_ref] [bib_ref] Histological types of lung cancer among smelter workers exposed to arsenic, Pershagen [/bib_ref] [bib_ref] Arsenic and cigarette smoke synergistically increase DNA oxidation in the lung, Hays [/bib_ref] [bib_ref] Case-control study of bladder cancer and exposure to arsenic in Argentina, Bates [/bib_ref] [bib_ref] Case-control study of bladder cancer and drinking water arsenic in the western..., Steinmaus [/bib_ref] [bib_ref] Modification of risk of arsenic-induced skin lesions by sunlight exposure, smoking, and..., Chen [/bib_ref] [bib_ref] The risk of arsenic induced skin lesions in Bangladeshi men and women..., Lindberg [/bib_ref] [bib_ref] Methylation study of a population environmentally exposed to arsenic in drinking water, Hopenhayn-Rich [/bib_ref] [bib_ref] Gender and age differences in the metabolism of inorganic arsenic in a..., Lindberg [/bib_ref]
# Materials and methods
Study population. This study is part of a population-based case-referent study concerning the development of skin lesions in relation to arsenic exposure via drinking water carried out in Matlab, a rural area 53 km southeast of Dhaka, Bangladesh [bib_ref] Arsenic exposure and age-and sex-specific risk for skin lesions: a population-based casereferent..., Rahman [/bib_ref] [bib_ref] Prevalence of arsenic exposure and skin lesions. A population based survey in..., Rahman [/bib_ref]. More than 60% of the tube wells in this area have concentrations above 50 µg/L, which is the Bangladeshi standard for drinking Background: We recently reported that the main reason for the documented higher prevalence of arsenic-related skin lesions among men than among women is the result of less efficient arsenic metabolism. oBjective: Because smoking has been associated with less efficient arsenic methylation, we aimed to elucidate interactions between tobacco use and arsenic metabolism for the risk of developing skin lesions. Methods: We used a population-based case-referent study that showed increased risk for skin lesions in relation to chronic arsenic exposure via drinking water in Bangladesh and randomly selected 526 of the referents (random sample of inhabitants > 4 years old; 47% male) and all 504 cases (54% male) with arsenic-related skin lesions to measure arsenic metabolites [methylarsonic acid (MA) and dimethylarsinic acid (DMA)] in urine using high-performance liquid chromatography (HPLC) and inductively coupled plasma mass spectrometry (ICPMS). results: The odds ratio for skin lesions was almost three times higher in the highest tertile of urinary %MA than in the lowest tertile. Men who smoked cigarettes and bidis (locally produced cigarettes; 33% of referents, 58% of cases) had a significantly higher risk for skin lesions than did nonsmoking men; this association decreased slightly after accounting for arsenic metabolism. Only two women smoked, but women who chewed tobacco (21% of referents, 43% of cases) had a considerably higher risk of skin lesions than did women who did not use tobacco. The odds ratio (OR) for women who chewed tobacco and who had ≤ 7.9 %MA was 3.8 [95% confidence interval (CI), 1.4-10] compared with women in the same MA tertile who did not use tobacco. In the highest tertile of %MA or %inorganic arsenic (iAs), women who chewed tobacco had ORs of 7.3 and 7.5, respectively, compared with women in the lowest tertiles who did not use tobacco. conclusion: The increased risk of arsenic-related skin lesions in male smokers compared with nonsmokers appears to be partly explained by impaired arsenic methylation, while there seemed to be an excess risk due to interaction between chewing tobacco and arsenic metabolism in women. water, and 70% are above the WHO maximum guideline of 10 µg As/L. In Matlab, the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B) has been running a comprehensive Health and Demographic Surveillance System (HDSS) that covers a population of 220,000.
We obtained informed consent from all participants, and the study was approved by both the ICDDR,B Ethical Review Committee and the ethics committee at the Karolinska Institutet in Stockholm. Mitigating activities such as painting wells with elevated arsenic concentrations and installing filters were initiated as described elsewhere [bib_ref] Sustainable safe water options in Bangladesh: experiences from the arsenic project in..., Jakariya [/bib_ref] [bib_ref] Prevalence of arsenic exposure and skin lesions. A population based survey in..., Rahman [/bib_ref].
Cases and referents recruitment. The recruitment of persons with skin lesions (504 cases) and the selection of referents (1,830 referents) are described elsewhere [bib_ref] Arsenic exposure and age-and sex-specific risk for skin lesions: a population-based casereferent..., Rahman [/bib_ref]. In short, all residents > 4 years of age who had lived in the study area for at least 6 months were eligible for the present study (180,811 individuals). In total, 166,934 (92%) of the eligible individuals were interviewed and examined for skin lesions by well-trained field teams, and 1,682 suspected cases were referred to the study physicians (one male and one female) at the health centers. Eventually, 504 cases were diagnosed with arsenic-induced skin lesions (defined as hyperpigmentation, hypopigmentation, and keratosis). For details regarding the ascertainment of cases, see [bib_ref] Arsenic exposure and age-and sex-specific risk for skin lesions: a population-based casereferent..., Rahman [/bib_ref]. All cases provided urine samples. Referents were randomly selected from the HDSS database with the criteria of more than 4 years of age, living in the area for at least 6 months, and drinking water from the area at least once a week. Selected referents were interviewed and invited to attend the clinic to be examined for skin lesions by a physician and to provide urine samples. A total of 1,579 referents attended the clinic.
Data collection. The field teams interviewed all individuals about their history of water consumption and the water sources used, including location, during each calendar year since 1970 or since birth if later than 1970, as described previously in more detail [bib_ref] Prevalence of arsenic exposure and skin lesions. A population based survey in..., Rahman [/bib_ref]. Data on socioeconomic status (SES) were collected from the HDSS database and were defined in terms of assets relevant to these rural settings [bib_ref] Prevalence of arsenic exposure and skin lesions. A population based survey in..., Rahman [/bib_ref]. Information on tobacco use, obtained in the interviews, was divided into cigarette smoking, bidi (locally produced cigarettes) smoking, or chewing tobacco, the latter consisting of dried tobacco leaves or zarda (a type of chewing tobacco, often used with areca nut, slaked lime, and betel leaves). Water samples were collected from all functional wells in the area and stored at -20°C (n = 13,286). Arsenic concentrations in drinking water were determined using atomic absorption spectrometry with hydride generation, with addition of hydrochloric acid and potassium iodine combined with heating [bib_ref] A modified routine analysis of arsenic content in drinking-water in Bangladesh by..., Wahed [/bib_ref]. For samples with concentrations below the limit of detection (LOD; 1.0 µg/L), half the LOD was used in the calculations.
Arsenic exposure. Both the average and the cumulative historical arsenic exposure were calculated as the time-weighted mean arsenic concentration of drinking water of all sources used since 1970 or birth. The cumulative arsenic exposure was calculated by summing the arsenic concentration multiplied by the number of years of usage (micrograms per liter × years) for all water sources used since 1970. From the interviews of the participants regarding their water consumption history, we were able to collect data on which year the participant started to drink well water (after 1970) and thereby the age at first exposure to tube well water. Very few wells were constructed before 1970, at which time the registration of the wells in the HDSS began.
Urine arsenic measures. Spot urine samples were collected in 20 mL polyethylene containers and stored at -20°C. We randomly selected 526 samples from all referents and all 504 cases for analysis of arsenic metabolites in urine for evaluation of the individual arsenic methylation efficiency. The individual pattern of arsenic metabolites in urine is shown to be fairly stable over time [bib_ref] Intra-individual variation in the metabolism of inorganic arsenic, Concha [/bib_ref] [bib_ref] Variability in biomarkers of arsenic exposure and metabolism in adults over time, Kile [/bib_ref] [bib_ref] Intraindividual variability in arsenic methylation in a U.S. population, Steinmaus [/bib_ref]. Speciation analysis of arsenic metabolites in urine was performed by an inductively coupled plasma mass spectrometer (ICP-MS; Agilent 7500ce, Agilent Technologies, Waldbronn, Germany) together with an Agilent 1100 chromatographic system equipped with solvent degasser, auto sampler, and temperature-controlled column. For the separation of trivalent arsenic [As(III)], MA, DMA, and pentavalent arsenic [As(V)], a Hamilton PRP-X100 anion exchange column (4.6 mm × 250 mm) was used [bib_ref] Arsenic exposure in Hungary, Romania and Slovakia, Lindberg [/bib_ref] [bib_ref] Evaluation of the three most commonly used analyti-Evaluation of the three most..., Lindberg [/bib_ref]. For quality control, we used a Japanese reference urine certified for arsenic [bib_ref] Human urine certified reference material for arsenic speciation, Yoshinaga [/bib_ref] and a spiked urine sample. Mean concentrations of As(III), DMA, MA, and As(V) in the reference urine were 3.9 ± 0.4, 43 ± 3, 3.0 ± 0.4, and 0.1 ± 0.1 (n = 79 during an 8-week period), respectively. The mean concentrations of the spiked urine samples of As(III), DMA, MA, and As(V) were 1.5 ± 0.6, 58 ± 3, 10 ± 0.6, and 16 ± 1.3 (n = 82 during an 8-week period), respectively. The results of interlaboratory comparisons are described in more detail elsewhere [bib_ref] Evaluation of the three most commonly used analyti-Evaluation of the three most..., Lindberg [/bib_ref]. Arsenic concentrations in urine were adjusted to the average specific gravity in this population (1.012 g/cm 3 ), as measured by a refractometer (Uricon-Ne, ATAGO Co. Ltd, Tokyo, Japan) to adjust for variation in dilution in the urine samples .
We also measured the arsenic concentrations of various forms of cigarettes, using ICP-MS after microwave-assisted acid digestion at high temperature and pressure as previously described for blood and breast milk [bib_ref] Breast-feeding protects against arsenic exposure in Bangladeshi infants, Fangstrom [/bib_ref].
Statistical analyses. We used SPSS 14.0 for Windows (SPSS Inc., Chicago, IL, USA) to perform all statistical analyses. Multivariate logistic regression analyses were performed to estimate the odds ratios (ORs) and corresponding confidence intervals (CIs) for having skin lesions at different proportions of arsenic metabolites in urine. Continuous variables were stratified into tertiles when estimating the ORs for having skin lesions. All multi variate associations were simultaneously adjusted for sex (man/woman), age (continuous), SES (continuous), tobacco use (no/yes), and cumulative arsenic exposure (continuous). We chose to adjust all models for cumulative arsenic exposure instead of average arsenic exposure or urinary arsenic concentrations. The cumulative arsenic exposure influenced the associations slightly more than did the average arsenic exposure, and adjusting for the urinary arsenic concentrations did not make any difference in the models when tested (data not shown). We calucated p for trend by treating the cate gorical variables as continuous variables in the models. To evaluate potential biologic interactions between the different risk factors, that is, arsenic metabolism and smoking, on arsenic-related skin lesions, we calculated the relative excess risk due to additive interactions (RERI) and corresponding CIs . These calculations were performed according to [bib_ref] Calculating measures of biological interaction, Andersson [/bib_ref] , using an Excel sheet (EpiNET 2008). We used p-values < 0.05 for statistical significance.
# Results
Sex-specific characteristics of the cases and referents are presented in [fig_ref] Table 1: ORs [/fig_ref]. The cases were more often men than women (54% vs. 46%) and older than the referents (overall, 40 vs. 31 years old). The cases also had higher SES, as measured by household asset scores, and they smoked cigarettes or bidis (men only) or used zarda more often than did the referents. The three different measures of arsenic exposure, that is, cumulative lifetime exposure to arsenic in drinking water (micrograms per liter × years), average lifetime exposure to arsenic (micrograms per liter), and current total exposure to iAs, as measured by the concentration of arsenic metabolites in urine, are also presented in [fig_ref] Table 1: ORs [/fig_ref]. The cases had significantly higher lifetime cumulative arsenic exposure and average lifetime arsenic exposure than did the referents but not higher urinary arsenic concentrations.
The number of years of tobacco use and the number of tobacco products used per day among cases and referents are shown in Supplemental Material, Table S1 (doi:10.1289/ ehp.0900728). The pattern of tobacco use differed markedly between men (46% smokers, 11% chewing tobacco) and women [only two female smokers, 31% chewing tobacco (see Supplemental Material, [fig_ref] Table 2: Pattern of arsenic metabolites [/fig_ref] ]. We analyzed men and women separately. The associations between the different forms of tobacco use and urinary arsenic metabolites are shown in [fig_ref] Table 2: Pattern of arsenic metabolites [/fig_ref]. All forms of tobacco use were associated with an increased percentage of MA and a decreased percentage of DMA, whereas the percentage of iAs and the urinary arsenic concentration did not vary much by tobacco use.
Men who smoked cigarettes or bidis had significantly higher risk for skin lesions than did men who did not use tobacco (OR = 1.8; 95% CI, 1.1-3.1; adjusted for age, SES, and cumulative arsenic exposure; [fig_ref] Table 3: ORs for arsenic-related skin lesions in relation to tobacco use among the... [/fig_ref]. We observed a marked drop in the OR after adjusting for age, SES, and cumulative arsenic exposure (crude OR = 3.2). Entering the different covariables independently showed that age was the main confounding factor. The adjusted OR in men chewing tobacco was not significantly increased (adjusted OR = 0.80; 95% CI, 0.36-1.7; [fig_ref] Table 3: ORs for arsenic-related skin lesions in relation to tobacco use among the... [/fig_ref] ; however, the number of individuals was low. Because very few women smoked, we could not determine the effect on the risk for skin lesions among women. The multivariable-adjusted model that included all women who used tobacco (mainly chewing tobacco) showed that they had considerably higher prevalence of skin lesions than did women who did not use tobacco (adjusted OR = 2.4; 95% CI, 1.4-3.9; [fig_ref] Table 3: ORs for arsenic-related skin lesions in relation to tobacco use among the... [/fig_ref]. Excluding the two women who smoked cigarettes did not change the associations. To test for the influence of arsenic methylation on the association between tobacco use and arsenic-related skin lesions, we also adjusted the OR values for %MA in urine. As shown in [fig_ref] Table 3: ORs for arsenic-related skin lesions in relation to tobacco use among the... [/fig_ref] , the OR for men who smoked changed from 1.8 to 1.4 (95% CI, 0.80-2.4) after additional adjustment for %MA. Compared with women who did not use any tobacco, the OR for women who chewed tobacco changed from 2.4 to 2.0 (95% CI, 1.2-3.4) after adjusting for %MA and remained significantly increased.
To further evaluate interactions between smoking and arsenic metabolism on the risk for skin lesions, we stratified both tobacco use and urinary arsenic metabolites and analyzed the joint effects. The increasing risk for skin lesions with increasing proportion of MA and decreasing proportion of DMA was more pronounced among nontobacco using men and women [fig_ref] Table 4: Joint effect of arsenic metabolites in urine and tobacco use for the... [/fig_ref]. The OR for men who used tobacco (mostly smokers, because few chewed tobacco) with ≤ 7.9 %MA was 1.8 (95% CI, 0.53-6.3; p = 0.34), compared with men in the same tertile of %MA who did not use tobacco. Men within the highest tertile of %MA who used tobacco had an OR of 4.4 (95% CI, 1.7-11). This OR was essentially the same as the sum of the OR for non tobacco users in the highest %MA tertile and that for tobacco users in the lowest %MA tertile minus the baseline OR (OR values 3.8 + 1.8 -1.0 = 4.6). Similar results were obtained for %iAs; the OR joint was 2.8, which is approximately equal to the sum (minus 1) of the OR for nonsmokers in the highest %iAs tertile (OR 1.6) and that for smokers in the lowest %iAs (OR = 1.9). Among women, the OR for tobacco users (mainly chewing tobacco) with ≤ 7.9 %MA in urine was 3.8 (95% CI, 1.4-10), compared with nonusers of tobacco within the same tertile (p = 0.009; [fig_ref] Table 5: Joint effect of arsenic metabolites in urine and tobacco use for the... [/fig_ref]. Women in the highest %MA tertile who chewed tobacco had an OR of 7.3 (95% CI, 3.4-15), compared with women in the lowest %MA group who did not use tobacco. This OR was higher than the predicted additive risks (5.7), although the RERI was not statistically significant (1.5; 95% CI, -3.7 to 6.7). Similarly, the OR joint for women in the highest %iAs tertile who used tobacco was 7.5, which was higher than the predicted additive risks (2.5). The RERI was 5.0 (95% CI, -1.3 to 11.4), the attributable proportion due to interaction was 0.67 (0.36-0.98), and the synergy index was 4.4 (1.2-16), which indicated a biologic interaction.
# Discussion
This population-based case-referent study in Bangladesh is the first to evaluate the combined effects of arsenic exposure, arsenic metabolism, and use of tobacco for the risk of arsenic-related skin effects. All forms of tobacco use were associated with less efficient methylation of arsenic. Among men, there appeared to be an additive effect of poor arsenic methylation (high iAs and high %MA) and smoking for the development of arsenic-induced skin lesions, although a high %MA increased the risk more than did smoking. Because very few women smoked cigarettes or bidis, an interaction between arsenic methylation and smoking in women could not be evaluated. Another new finding in the present study was that tobacco chewing, which is much more common among Bangladeshi women than smoking, was also a risk factor for developing arsenic-related skin lesions in women. The high ORs for skin lesions among the women who chewed tobacco in the highest tertiles of %iAs or %MA (7.5 and 7.3, respectively), compared with nontobacco using women with efficient arsenic methylation, suggest an interaction, although the RERI values were not quite significant. For men, an association between chewing tobacco and skin lesions was observed in the crude analysis only, but the sample size was small and the CIs wide. Further studies on larger cohorts are warranted for firm conclusions concerning the biologic interactions between various tobacco use, arsenic exposure, and arsenic metabolism. In any case, the use of various forms of tobacco should be considered in the risk assessment of arsenic and in the comparison of arsenic-related health risks among populations. Tobacco smoking has been identified as an independent risk factor of non melanoma skin cancer [bib_ref] Relation between smoking and skin cancer, De Hertog [/bib_ref] [bib_ref] The effect of solar UVB doses and vitamin D production, skin cancer..., Grant [/bib_ref] [bib_ref] A prospective study of incident squamous cell carcinoma of the skin in..., Grodstein [/bib_ref] , psoriasis [bib_ref] Smoking and the risk of psoriasis in women: Nurses' Health Study II, Setty [/bib_ref] , and premature skin aging [bib_ref] Environmental Health Perspectives Effect of smoking on skin elastic fibres: morphometric and..., Just [/bib_ref] [bib_ref] Tobacco smoke causes premature skin aging, Morita [/bib_ref] , but few studies have investigated the modifying effect of smoking on the arsenic-related hyper pigmentation and hyperkeratosis. [bib_ref] Modification of risk of arsenic-induced skin lesions by sunlight exposure, smoking, and..., Chen [/bib_ref] reported a significant synergistic effect between the highest level of arsenic exposure via drinking water (> 113 µg/L) and tobacco smoking for the risk of skin lesions among Bangladeshi men, but much weaker interaction effects among women. These authors suggested that the interaction could be due to immuno suppression caused by tobacco smoking, inhibition of arsenic methylation, or the prevalent smoking of bidis, the filterless, locally produced cigarettes with raw tobacco. Bidis are popular in rural areas in Bangladesh and are claimed to contain more carcinogenic substances than do cigarettes.
In the present study, the effect of smoking on arsenic-related skin lesions was studied only in men, because, by tradition, very few women in Bangladesh are smokers. According to personal interviews, almost half the men were smokers, whereas only two of more than 500 women smoked. It is highly unlikely that we have any differential misclassification in the data on tobacco use. Smoking or other forms of tobacco use are not linked to any stigma and are in no way considered to be associated with the arsenic-induced skin lesions by the study population. The increased risk of skin lesions among men who smoked was not due to additional exposure to arsenic via smoking. The concentrations of arsenic in different brands of cigarettes from local shops in Matlab ranged between 0.13 and 0.29 µg/g (mean 0.21 µg/g, n = 5) and between 0.24 and 0.27 µg/g (mean 0.25 µg/g, n = 3) for bidis. Thus, it could be estimated that the inhaled amount of arsenic by smoking 10 cigarettes or bidis/day, for example, was about 2 µg. Even though a considerable part of this arsenic is absorbed in the lungs, the arsenic uptake from smoking is negligible compared with that from drinking water (70% of the wells had > 10 µg As/L; [bib_ref] Prevalence of arsenic exposure and skin lesions. A population based survey in..., Rahman [/bib_ref] and food [bib_ref] Gender and age differences in the metabolism of inorganic arsenic in a..., Lindberg [/bib_ref].
Instead, we found that smoking was associated with higher %MA in urine, which is a known risk factor for arsenic-related skin effects [bib_ref] Arsenic metabolism, genetic susceptibility, and risk of premalignant skin lesions in Bangladesh, Ahsan [/bib_ref] [bib_ref] Arsenic methylation and skin cancer risk in southwestern Taiwan, Chen [/bib_ref] [bib_ref] Altered profile of urinary arsenic metabolites in adults with chronic arsenicism. A..., Razo [/bib_ref] [bib_ref] Serum beta-carotene level, arsenic methylation capability, and incidence of skin cancer, Hsueh [/bib_ref] [bib_ref] Gender and age differences in the metabolism of inorganic arsenic in a..., Lindberg [/bib_ref] [bib_ref] Arsenic methylation capacity and skin cancer, Yu [/bib_ref]. Elevated %MA in urine may be related to the highly toxic trivalent intermediate metabolite MA(III) [bib_ref] Impact of trivalent arsenicals on selenoprotein synthesis, Ganyc [/bib_ref] [bib_ref] Differential effects of trivalent and pentavalent arsenicals on cell proliferation and cytokine..., Vega [/bib_ref] in the tissues, including skin [bib_ref] Mechanisms of arsenic biotransformation, Vahter [/bib_ref]. The decrease in the second step in the methylation of arsenic (to DMA) by smoking is in agreement with previous findings [bib_ref] Methylation study of a population environmentally exposed to arsenic in drinking water, Hopenhayn-Rich [/bib_ref] [bib_ref] The risk of arsenic induced skin lesions in Bangladeshi men and women..., Lindberg [/bib_ref] ; however, the mechanism by which this occurs is not clear. It may be that smoking inhibits the specific AS3MT involved in arsenic methylation or impairs one-carbon metabolism in general. Cigarette smoking is known to increase serum homocysteine concentration [bib_ref] Smoking and plasma homocysteine, O'callaghan [/bib_ref] [bib_ref] The Hordaland Homocysteine Study: a community-based study of homocysteine, its determinants, and..., Refsum [/bib_ref] , which, via the concurrent accumulation of S-adenosylhomocysteine, exerts a strong inhibition of S-adenosylmethionine-dependent transmethylation reactions, including those of arsenic [bib_ref] Folate, homocysteine, and arsenic metabolism in arsenic-exposed individuals in Bangladesh, Gamble [/bib_ref] [bib_ref] The effect of methyltransferase inhibition on the metabolism of [74As]arsenite in mice..., Marafante [/bib_ref]. Smokers also tend to have lower levels of folate and vitamin B6 and B12 [bib_ref] Smoking and plasma homocysteine, O'callaghan [/bib_ref] , all of which are essential for homocysteine metabolism. The smoking-related increase in homocysteine is most likely less pronounced in women, in whom the estrogen-dependent upregulation of endogenous choline synthesis may, via oxidation to betaine, contribute to remethylation of homocysteine [bib_ref] Gene response elements, genetic polymorphisms and epigenetics influence the human dietary requirement..., Zeisel [/bib_ref]. Indeed, the study by [bib_ref] Modification of risk of arsenic-induced skin lesions by sunlight exposure, smoking, and..., Chen [/bib_ref] observed a much stronger effect of smoking on arsenic-related skin lesions in men than in women.
Another new finding in this study is that chewing tobacco is a risk factor for arsenicrelated skin effects for women and that there appears to be an interaction with poor metabolism of arsenic. For men, the effect of chewing tobacco on the risk of skin lesion seems to be much less, but the number of cases is too small for a firm conclusion. About 31% of the women reported chewing tobacco, which locally is called shada (dried tobacco leaves) or zarda (processed tobacco leaves in a paste) [bib_ref] Sociodemographic characteristics of tobacco consumers in a rural area of Bangladesh, Choudhury [/bib_ref]. In zarda, the tobacco is often mixed with sliced areca nut, lime, and sometimes a leaf of the piper betel plant, and the adverse effects may be caused by this mixture rather than the tobacco alone. Areca nut, the seed of Areca catechu, is used in a variety of chewed products, often mixed with tobacco or betel leaves. Both betel quid and areca nut have been associated with increased risk for oral epithelial malignancy [bib_ref] The upregulation of heat shock protein 70 expression in areca quid chewingassociated..., Lee [/bib_ref]. In a previous study in reported that betel nut use, but not tobacco chewing or cigarette smoking, was associated with skin lesions. As that study matched by sex, it is not known if the associations were sex dependent. In the present study, zarda contained about half a microgram of arsenic per gram of tobacco (0.33-0.54 µg/g; mean 0.45 µg/g; n = 8), which is about twice as much as in cigarettes and bidis. Still, the arsenic exposure from chewing zarda was low compared with that from the drinking water and food.
[table] Table 1: ORs (95% CIs) for skin lesions by characteristics of case-referent participants (504 cases, 528 referents). [/table]
[table] Table 2: Pattern of arsenic metabolites (percentages of iAs, MA, DMA) in urine, by tobacco use and sex. [/table]
[table] Table 3: ORs for arsenic-related skin lesions in relation to tobacco use among the 528 referents and 504 skin lesion cases, stratified by sex. [/table]
[table] Table 4: Joint effect of arsenic metabolites in urine and tobacco use for the risk of arsenic-related skin lesions among men. [/table]
[table] Table 5: Joint effect of arsenic metabolites in urine and tobacco use for the risk of arsenic-related skin lesions among women. Tertiles of referents. b OR for the groups stratified by level of arsenic metabolites. c Adjusted for age, SES, and cumulative arsenic exposure. d Mainly tobacco chewing; only two women smoked. *p < 0.05. **p < 0.01. [/table]
|
ClC Channels and Transporters: Structure, Physiological Functions, and Implications in Human Chloride Channelopathies
The discovery of ClC proteins at the beginning of the 1990s was important for the development of the Cl − transport research field. ClCs form a large family of proteins that mediate voltage-dependent transport of Cl − ions across cell membranes. They are expressed in both plasma and intracellular membranes of cells from almost all living organisms. ClC proteins form transmembrane dimers, in which each monomer displays independent ion conductance. Eukaryotic members also possess a large cytoplasmic domain containing two CBS domains, which are involved in transport modulation. ClC proteins function as either Cl − channels or Cl − /H + exchangers, although all ClC proteins share the same basic architecture. ClC channels have two gating mechanisms: a relatively well-studied fast gating mechanism, and a slow gating mechanism, which is poorly defined. ClCs are involved in a wide range of physiological processes, including regulation of resting membrane potential in skeletal muscle, facilitation of transepithelial Cl − reabsorption in kidneys, and control of pH and Cl − concentration in intracellular compartments through coupled Cl − /H + exchange mechanisms. Several inherited diseases result from C1C gene mutations, including myotonia congenita, Bartter's syndrome (types 3 and 4), Dent's disease, osteopetrosis, retinal degeneration, and lysosomal storage diseases. This review summarizes general features, known or suspected, of ClC structure, gating and physiological functions. We also discuss biophysical properties of mammalian ClCs that are directly involved in the pathophysiology of several human inherited disorders, or that induce interesting phenotypes in animal models.
# Introduction
Ion transporters typically use the electrochemical gradient of one substrate (or another source of energy such as ATP) to transport another substrate in a well-defined stoichiometry and direction. This is a relatively slow process limited by the number of ions/substrate that can bind to the transporter in a given transport cycle, as well as by the need for conformational changes to deliver the transported substrate to the opposing side of the membrane. Conversely, ion channels passively move ions down electrochemical gradients at a high-rate flux through a pore with a defined selectivity.
Ion channels commonly exist in four states: closed, open, inactive, and desensitized, with each state having a different ion conductance. "Gating" is the term used to describe changes between the different states. Various factors, including voltage, ligand binding, second messengers, volume, and temperature modulate ion channel gating. Because of their important role in cell function, mutations in ion channel genes that cause impaired channel function are associated with a variety of human diseases, generally known as channelopathies. Channelopathies affect the nervous, cardiovascular, respiratory, endocrine, urinary, and immune systems.
Many years of intense research have focused on how changes in ion channels' biophysical properties can induce drastic physiological changes at the cellular and tissular levels, subsequently causing severe and even lethal human diseases. The ultimate goal of this research is the development of specific targeted pharmacotherapies to treat channelopathies. Perhaps the most well-known example of such a therapy is the treatment of cystic fibrosis patients with corrector and potentiator drugs that specifically target the Cystic Fibrosis Transmembrane conductance Regulator (CFTR) chloride channel to alleviate mutations affecting its trafficking, folding, and function. Although this extraordinary translational development is still in its infancy, with less than 4 years of clinical use, it has required over two decades of research to reach the necessary level of understanding of the biophysical and functional parameters of the CFTR chloride channel.
Prior to cloning of the first chloride channels (ClC-0 and CFTR), chloride channels were of little interest to scientists, as Cl − was considered to be in electrochemical equilibrium across cell membranes. When studying action potentials, cation channels (Na + , K + , and Ca 2+ ) were considered the major players; Cl − flux was seen as a mere nuisance. Chloride movement across membranes can change both the concentration of the substrate, Cl − , and the electrical charge between the compartments. As an electronegative ion, chloride plays an important role in regulating the excitability of neurons and muscles through changes in the membrane potential. In epithelia, the Cl − concentration gradient drives the direction of ion movement through ion transporters, which helps maintain intraand extra-cellular osmotic homeostasis. The cloning of the CFTR chloride channel [bib_ref] Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA, Riordan [/bib_ref] and the Torpedo electric organ voltage-gated chloride channel ClC-0 [bib_ref] Primary structure of Torpedo marmorata chloride channel isolated by expression cloning in..., Jentsch [/bib_ref] were important breakthroughs in chloride channel research, paving the way for subsequent high impact publications on chloride transport. ClC-0 was not characteristic of any other chloride transporter previously described, and thus became the first member of the new ClC chloride channel family. Nine mammalian ClC proteins have been identified since the discovery of ClC-0. Four of these ClCs are expressed in the plasma membrane and operate as channels [fig_ref] TABLE 1 |: Mammalian ClC chloride channels [/fig_ref] , and the other five are Cl − /H + exchangers localized to intracellular membranes [fig_ref] TABLE 2 |: Mammalian ClC chloride exchangers [/fig_ref]. The ClC family of Cl − transporters is the focus of this review.
## The clc family
The discovery of Torpedo ClC-0 channel by [bib_ref] Primary structure of Torpedo marmorata chloride channel isolated by expression cloning in..., Jentsch [/bib_ref] garnered the attention of the scientific community toward the ClC protein family. ClC proteins occur in all phyla, with nine members present in mammals (ClC-1 to . Three of the ClC proteins contain a β-subunit (ClC-Ka, ClC-Kb, and ClC-7), which is essential for proper transport function, and another member (ClC-2) contains a non-essential β-subunit that changes its gating properties .
Although [bib_ref] Primary structure of Torpedo marmorata chloride channel isolated by expression cloning in..., Jentsch [/bib_ref] cloned the first ClC member, the Miller group discovered most of the surprising and unique properties of this family prior to this. Observations of single channel current recordings from Torpedo electroplax Cl − channels demonstrated an unusual gating behavior, with bursts containing two open conductance levels spaced by long periods of channel closing [bib_ref] Open-state substructure of single chloride channels from Torpedo electroplax, Miller [/bib_ref] [bib_ref] Single chloride channels from Torpedo electroplax. Activation by protons, Hanke [/bib_ref]. Since one conductance level was the double of the other, they assumed that the channel functions as a dimer, with each subunit having its own independent ion pathway (protopore). Gating of one or both subunits' protopores explained the two distinct conductance levels observed; meanwhile, the single closed state suggested that despite the two protopores functioning independently, some asyet-unknown mechanism closed them simultaneously [bib_ref] Steady-state coupling of ion-channel conformations to a transmembrane ion gradient, Richard [/bib_ref]. In this now well-established double-barrel model there is a fast gate (opening and closing events within a burst) occurring on a time scale of milliseconds and a slow (also called common) gate, in which both protopores are closed on a time scale of seconds, reflecting the single closed state observed in single channel analysis. All channel properties identified by Miller's group were afterward attributed to the ClC family of chloride channels.
In ClCs-unlike cation channels, whose gating is regulated by voltage sensors controlled by the membrane potentialthe permeant ion (Cl − ) is itself responsible for the voltagedependent gating, and protons influence the gating [bib_ref] Steady-state coupling of ion-channel conformations to a transmembrane ion gradient, Richard [/bib_ref] [bib_ref] Gating of the voltage-dependent chloride channel CIC-0 by the permeant anion, Pusch [/bib_ref] [bib_ref] How membrane proteins sense voltage, Bezanilla [/bib_ref]. That is, intra-and extra-cellular changes in Cl − concentration and pH modulate ClC channel function. In general, ClC channels have an anion selectivity sequence of Cl − > Br − > I − and are largely impermeable to cations [bib_ref] Permeation and block of the skeletal muscle chloride channel, ClC-1, by foreign..., Rychkov [/bib_ref].
In another surprising discovery, researchers have determined that while all ClC proteins share the same basic structure, some function as chloride-proton exchangers with a 2Cl − /1H + stoichiometry, instead of classical chloride channels (Accardi [bib_ref] Inactivation of muscle chloride channel by transposon insertion in myotonic mice, Steinmeyer [/bib_ref] ClC-2/ (GlialCAM) Brain; kidney; liver; heart; pancreas; skeletal muscles; lungs and GI tract Transepithelial transport Leukodystrophy, azoospermia Retinal and testes degeneration; leukodystrophy ClC-Ka/Barttin Inner ear; Kidney Transepithelial transport Loss of Barttin or both ClC-Ks: Bartter IV (renal salt loss and deafness Diabetes insipidus [bib_ref] Overt nephrogenic diabetes insipidus in mice lacking the CLC-K1 chloride channel, Matsumura [/bib_ref] ClC-Kb/Barttin Loss of ClC-Kb: Bartter III (renal salt loss)
The table illustrates tissue expression, function and pathologies related to dysfunction or absence of ClC channels from plasma membranes. Both ClC-K a and b isoforms require Barttin as an obligatory β-subunit for trafficking, stability and function. GlialCAM is a non-essential β-subunit of ClC-2 and, when associated, change ClC-2 localization and properties in glial cells. Retinal and brain degeneration [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref] [bib_ref] Altered GABAergic function accompanies hippocampal degeneration in mice lacking ClC-3 voltage-gated chloride..., Dickerson [/bib_ref] [bib_ref] CLC-3 deficiency leads to phenotypes similar to human neuronal ceroid lipofuscinosis, Yoshikawa [/bib_ref] ClC-4 Skeletal muscles; brain and heart Ion homeostasis of endosomes?
Intellectual disabilities?
ClC-5 Kidney; intestine Ion homeostasis of early endosomes Dent's disease Impaired renal endocytosis [bib_ref] Mice lacking renal chloride channel, CLC-5, are a model for Dent's disease,..., Wang [/bib_ref] ClC [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] The table illustrates tissue expression, function and pathologies related to defect or absence of the respective ClC exchangers from vesicles of the endosomal/lysosomal pathway. ClC-7 needs Ostm1 as an essential β-subunit to be stable and functional at lysosomal membranes. [bib_ref] Secondary active transport mediated by a prokaryotic homologue of ClC Cl-channels, Accardi [/bib_ref] [bib_ref] Chloride/ proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5, Picollo [/bib_ref] [bib_ref] Voltage-dependent electrogenic chloride/ proton exchange by endosomal CLC proteins, Scheel [/bib_ref]. In mammals, five ClC proteins function as Cl − /H + exchangers (ClC-3 to ClC-7) and are generally localized to intracellular membranes, while the other four (ClC-1, ClC-2, ClC-Ka, and ClC-Kb) function as bona fide chloride channels, strictly localized to the plasma membrane. Malfunctions in chloride conductance or Cl − /H + translocation are causes of genetically inherited diseases [bib_ref] Emerging roles of chloride channels in human diseases, Puljak [/bib_ref] [bib_ref] Chloride channelopathies, Planells-Cases [/bib_ref] [bib_ref] CLC channel function and dysfunction in health and disease, Stölting [/bib_ref].
## Clc protein structure
In 2002, high-resolution crystal structures of two bacterial ClC exchangers were resolved (EcClC from E. coli and StClC from S. typhimurium). Exhibiting a complex topology, each ClC subunit has 18 α-helices that are variable in length and remarkably tilted. Most of the α-helices fail to traverse the membrane and display an internal anti-parallel repeat architecture. This intriguing arrangement of helices makes it possible for residues from distant parts of the protein to come together at the center of the subunit, forming the ion selectivity filter for Cl − conductance [bib_ref] X-ray structure of a CLC chloride channel at 3.0 A reveals the..., Dutzler [/bib_ref] [bib_ref] Gating the selectivity filter in ClC chloride channels, Dutzler [/bib_ref].
In Dutzler's StClC structure, each ClC subunit has three highly conserved Cl − binding sites, which feature a partial positive charge formed by amino acid residues located in the N-terminal portion of specific α-helices (D, F, N, and R). In the crystal structure, Cl − could be found at three specific sites made up by these amino acids: (1) an internal site (S int ) in contact with the intracellular environment, (2) a central site (S cen ) buried in the membrane bilayer, and (3) an external site (S ext ) in contact with the extracellular solution. In this structure, S int and S cen are occupied by Cl − ions, whereas S ext is occupied by the negatively charged side-chain of a conserved glutamate (E148; helix F) named Glu ext . In S cen , Cl − ions are coordinated mainly by residues S107 (helix D) and Y445 (helix R), also called Ser cen and Tyr cen , respectively. A Cl − ion occurs in S ext only following mutation or protonation of E148, which renders ClC gating proton-dependency. Importantly, mutation of this glutamate residue (E148Q, which mimics protonation of the carboxylate side chain) abolishes voltage and chloride-dependent gating in ClC channels and uncouples Cl − /H + exchange, turning the proteins into passive chloride conductors [bib_ref] Gating the selectivity filter in ClC chloride channels, Dutzler [/bib_ref] [bib_ref] Secondary active transport mediated by a prokaryotic homologue of ClC Cl-channels, Accardi [/bib_ref]. E148 has been termed the 'gating glutamate, ' given its essential role in ClC protein function.
Some researchers have proposed that Cl − and E148 compete for S ext, and that Cl − conductance (during the pore opening) occurs only when the side-chain of E148 is displaced from S ext by extracellular Cl − [bib_ref] Coupling gating with ion permeation in ClC channels, Chen [/bib_ref]. Presumably, this is the reason that ClC gating is dependent on extracellular Cl − concentration. While the 'gating glutamate' in the S ext is suggested to be the molecular determinant of protopore gating [bib_ref] Gating the selectivity filter in ClC chloride channels, Dutzler [/bib_ref] , S107 in the S cen is thought to contribute to Cl − selectivity, as mutation of this residue to proline changes anion selectivity to NO 3 − [bib_ref] Conversion of the 2Cl-/1H+ antiporter ClC-5 in a NO3-/H+ antiporter by a..., Zifarelli [/bib_ref]. S int is located close to where the intracellular solution bathes the selectivity filter, and residues in helix D coordinate Cl − ions in this position [bib_ref] Gating the selectivity filter in ClC chloride channels, Dutzler [/bib_ref].
For ClC exchangers to function, a proton pathway is also required, although there is currently no consensus on how protons cross the transport pathway. A glutamate residue (E203), located at the intracellular interface (named Glu int ) is suggested to be the proton acceptor coupling H + and Cl − transport, as mutation of this residue abolishes proton transport [bib_ref] Separate ion pathways in a Cl-/H+ exchanger, Accardi [/bib_ref]. Glu ext is conserved in both channels and exchangers and is involved in both Cl − and H + conductance, whereas Glu int is only conserved in exchangers and participates only in H + transport [bib_ref] Secondary active transport mediated by a prokaryotic homologue of ClC Cl-channels, Accardi [/bib_ref] [bib_ref] Separate ion pathways in a Cl-/H+ exchanger, Accardi [/bib_ref]. Concurrent mutation of the intracellular and extracellular glutamates leads to a loss of proton transport, although Cl − transport is still active. Glu int localizes away from the Cl − selectivity filter, in a region closer to the subunit's interface. Although experimental data is lacking, Glu int and Glu ext appear to cooperate to facilitate proton transport. In the proposed mechanism, Glu int accepts a H + from one side of the membrane and transfers it to Glu ext , which then completes the translocation process [bib_ref] Separate ion pathways in a Cl-/H+ exchanger, Accardi [/bib_ref]. However, it is not clear how protons would traverse the gap between Glu int and Glu ext , and because of Glu int localization, the pathways for Cl − and H + would diverge in the intracellular side converging only in the extracellular side, at Glu ext .
The first relatively high-resolution structure of a mammalian ClC channel (a bovine ClC-K) was solved by cryo-electron microscopy [bib_ref] Structure of a CLC chloride ion channel by cryo-electron microscopy, Park [/bib_ref]. Bovine ClC-K (henceforth, bClC-K) shares 84% sequence similarity with human ClC-K channels and is only functional when co-expressed with the β-subunit barttin. bClC-K contains a valine residue (V166) substituted for Glu ext, which causes the channel to have a linear current-voltage relationship [bib_ref] Structure of a CLC chloride ion channel by cryo-electron microscopy, Park [/bib_ref]. Based on sequence homology the structure of bClC-K is predicted to be similar to other ClC family members [bib_ref] X-ray structure of a CLC chloride channel at 3.0 A reveals the..., Dutzler [/bib_ref] [bib_ref] Gating the selectivity filter in ClC chloride channels, Dutzler [/bib_ref] [bib_ref] Structure of a eurkaryotic CLC transporter defines an intermediate state in the..., Feng [/bib_ref]. However, the high-resolution structure reported by [bib_ref] Structure of a CLC chloride ion channel by cryo-electron microscopy, Park [/bib_ref] suggests some marked differences. bClC-K contains two extracellular loops, one connecting helices K and M and the other connecting helices I and J. Both loops are located at the extracellular entrance of the chloride pathway with the latter loop in close proximity to the Cl − selectivity filter. There is also a cytosolic loop connecting helices C and D that displays a unique conformation from ClC transporters. In bClC-K the loop contains Ser cen (S121 in ClC-K), which faces the cytosolic side, whereas in ClC transporters Ser cen is found facing S cen interacting with a Cl − ion together with Tyr cen [bib_ref] Structure of a CLC chloride ion channel by cryo-electron microscopy, Park [/bib_ref]. Based on these structural differences between bClC-K and ClC transporters the authors propose a hypothesis for the different mechanisms underlying ClC channel and transporter conductance. They suggest that in transporters Ser cen and Tyr cen form a kinect barrier (a constriction) in the middle of the Cl − pathway preventing Cl − leak uncoupled to H + during a transport cycle. In ClC channels they suggest that the unique positioning of Ser cen relieves the kinect barrier allowing higher Cl − conductance [bib_ref] Structure of a CLC chloride ion channel by cryo-electron microscopy, Park [/bib_ref].
The most well studied plant ClC protein is the anion/proton exchanger AtClC-a, from Arabidopsis thaliana. AtClC-a promotes the exchange of NO 3 rather than Cl − due to a substitution of Ser cen to proline [bib_ref] Conversion of the 2Cl-/1H+ antiporter ClC-5 in a NO3-/H+ antiporter by a..., Zifarelli [/bib_ref]. AtClC-a shows outwardly rectifying currents and a 2NO 3 /1H + stoichiometry when expressed in isolated vacuoles, similar to animals and prokaryotes ClC exchangers [bib_ref] The nitrate/proton antiporter AtCLCa mediates nitrate accumulation in plant vacuoles, De Angeli [/bib_ref]. Glu ext and Glu int are conserved in AtClC-a and inactivation of those residues produces similar effect as in the intracellular ClCs 4 and 5 [bib_ref] Residues important for nitrate/proton coupling in plant and mammalian CLC transporters, Bergsdorf [/bib_ref].
## Cytoplasmic domains
All eukaryotic ClCs (and some prokaryotic ClCs) have a large cytoplasmic domain involved in modulating the trafficking and function of ClC proteins [bib_ref] Functional and structural conservation of CBS domains from CLC chloride channels, Estévez [/bib_ref] [bib_ref] The role of the carboxyl terminus in ClC chloride channel function, Hebeisen [/bib_ref]. Mutations in the cytoplasmic domains cause severe defects in slow gating, and are also associated with human genetic diseases [bib_ref] Determinants of slow gating in ClC-0, the voltage-gated chloride channel of Torpedo..., Fong [/bib_ref] [bib_ref] Functional and structural conservation of CBS domains from CLC chloride channels, Estévez [/bib_ref] [bib_ref] Emerging roles of chloride channels in human diseases, Puljak [/bib_ref] [bib_ref] Chloride channelopathies, Planells-Cases [/bib_ref] [bib_ref] CLC channel function and dysfunction in health and disease, Stölting [/bib_ref]. The crystal structures of cytoplasmic domains from ClC-0, ClC-Ka, and ClC-5 have been resolved; cytoplasmic domains of each subunit contain two CBS domains that interact with one another via an extensive interface. The CBS domains also interact with the transmembrane part of the same subunit, and with CBS domains of the other subunit. Additionally, the cytoplasmic domains display a dimeric organization resembling the transmembranal architecture [bib_ref] Crystal structure of the cytoplasmic domain of the chloride channel ClC-0, Meyer [/bib_ref] [bib_ref] The structure of the cytoplasmic domain of the chloride channel ClC-ka reveals..., Markovic [/bib_ref] [bib_ref] Nucleotide recognition by the cytoplasmic domain of the human chloride transporter ClC-5, Meyer [/bib_ref] [bib_ref] Structure of a eurkaryotic CLC transporter defines an intermediate state in the..., Feng [/bib_ref] [bib_ref] Structure of a CLC chloride ion channel by cryo-electron microscopy, Park [/bib_ref].
The cytoplasmic domains connect with the α-helix R, which contains Tyr cen that participates directly in Cl − coordination during transport. As mutations in the cytoplasmic domains are involved in genetic diseases, several studies have addressed the influence of alterations in the cytoplasmic domains in channel gating behavior. A point mutation downstream of the second CBS domain (A885P) in ClC-1 results in a dramatic reduction in channel open probability at voltages near the optimal membrane potential for ClC-1 to function [bib_ref] Molecular basis for decreased muscle chloride conductance in the myotonic goat, Beck [/bib_ref]. Two truncated ClC-1 mutants (R875X and K894X), the first removing the whole region downstream of CBS2 and the second mimicking a naturally occurring mutation in myotonic patients, display changes in anion binding affinity, resulting in changes in the voltage dependence for both fast and slow gates [bib_ref] Carboxy-terminal truncations modify the outer pore vestibule of muscle chloride channels, Hebeisen [/bib_ref]. [bib_ref] Carboxy terminus splice variation alters ClC channel gating and extracellular cysteine reactivity, He [/bib_ref] analyzed two splice variants of Caenorhabditis elegans ClC channel, CLH3a and CLH3b, which display marked gating differences. CHL3a has a N-terminal splice insertion that when deleted do not alter gating properties. CHL3b has two splice insertions at the cytoplasmic domains, one between the two CBS domains and the second distal to CBS2. Deletion of either the insertion distal to CBS2 or the last 11 amino acids of CBS1 gives rise to channels with gating properties similar to CHL3a [bib_ref] Carboxy terminus splice variation alters ClC channel gating and extracellular cysteine reactivity, He [/bib_ref]. Those studies demonstrate that alterations at the cytoplasmic domains modify the conformation of the pore affecting channel gating.
Cytoplasmic domains of some ClC proteins interact with adenosine nucleotides. In ClC-1, binding of intracellular ATP inhibits the channel by stabilizing it in its closed state [bib_ref] Cytoplasmic ATP-sensing domains regulate gating of skeletal muscle ClC-1 chloride channels, Bennetts [/bib_ref]. ATP binding has the opposite effect in ClC-5 exchangers, activating the transporter. Binding of ATP to ClC-2 slows down the rate of activation and deactivation, but does not affect the maximal open probability of the channel. Nucleotides bind at the interface between the two CBS domains, as revealed by the crystal structure of ClC-5. The nucleotide binding site has no catalytic properties, and to it AMP, ADP, and ATP bind with equal affinity [bib_ref] Nucleotide recognition by the cytoplasmic domain of the human chloride transporter ClC-5, Meyer [/bib_ref]. There are no apparent nucleotide binding sites in the cytoplasmic domains of ClC-0 or ClC-Ka [bib_ref] Crystal structure of the cytoplasmic domain of the chloride channel ClC-0, Meyer [/bib_ref] [bib_ref] The structure of the cytoplasmic domain of the chloride channel ClC-ka reveals..., Markovic [/bib_ref]. Interestingly, the cytoplasmic domains of the plant ClC AtClC-a also interacts with adenosine nucleotides. At this exchanger, ATP reduces transport activity by a maximum of 60%. Unlike ClC-5, only ATP produces this effect, with AMP working only as a competitor limiting ATP inhibition when present in solution [bib_ref] ATP binding to the C terminus of the Arabidopsis thaliana nitrate/proton antiporter,..., De Angeli [/bib_ref].
The antiparallel dimerization observed with the CBS domains of ClC proteins following ATP binding is a feature also seen in the CFTR chloride channel, in which ATP binds at two conserved motifs at the interface of two intracellular nucleotide binding domains. These domains dimerize in a head-to-tail conformation leading to channel gating and chloride movement following conformational changes. By analogy, one may speculate that adenosine nucleotide binding to the CBS domains could cause protein rearrangements that affect channel behavior in some ClCs.
## Common gating
In contrast to the well-studied fast gating mechanism of ClC proteins, the molecular mechanism of the slow (common) gate is still obscure. Evidence suggests that extensive conformational rearrangements in the protein could contribute to the slow gate [bib_ref] Temperature dependence of fast and slow gating relaxations of ClC-0 chloride channels, Pusch [/bib_ref] [bib_ref] Zinc inhibits human ClC-1 muscle chloride channel by interacting with its common..., Duffield [/bib_ref] [bib_ref] Large movement in the C-terminus of CLC-0 chloride channel during slow gating, Bykova [/bib_ref] [bib_ref] Movement of hClC-1 C-termini during common gating and limits on their cytoplasmic..., Ma [/bib_ref]. Two facts support the idea that conformational changes promoted by the cytoplasmic domains may lead to the movement of critical transmembrane helices and play an important role in the common gating mechanism. First, point mutations at helices localized at the dimer interface cause changes in common gating [bib_ref] Involvement of helices at the dimer interface in ClC-1 common gating, Duffield [/bib_ref]. Second, the crystal structure of the eukaryotic ClC transporter show relevant connections between the CBS domains and helices H and I, localized at the dimer interface [bib_ref] Structure of a eurkaryotic CLC transporter defines an intermediate state in the..., Feng [/bib_ref].
Another interesting hypothesis suggests that ClC channels likely behave as 'broken' exchangers in which proton transport is involved in the common gating, suggesting that the conformational changes of channels' common gating and coupled Cl − /H + transport have an evolutionary linkage [bib_ref] The ClC-0 chloride channel is a ' broken' Cl-/H+ antiporter, Lísal [/bib_ref]. Since gating of ClC-0 channels is not in a thermodynamic equilibrium [bib_ref] Steady-state coupling of ion-channel conformations to a transmembrane ion gradient, Richard [/bib_ref] , the authors demonstrated that proton transport is involved in ClC-0 gating and is, in fact, the source of energy that keeps ClC channels in this asymmetric gating state [bib_ref] The ClC-0 chloride channel is a ' broken' Cl-/H+ antiporter, Lísal [/bib_ref]. Further support for this position comes from a study in which a small but reproducible H + transport demonstrated in ClC-1 channels was no longer identified in ClC-0 channels carrying the C212S mutation that abolish common gating [bib_ref] Chloride/ proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5, Picollo [/bib_ref].
There is evidence for a critical role for the Glu ext residue in this mechanism [bib_ref] Gating the selectivity filter in ClC chloride channels, Dutzler [/bib_ref] [bib_ref] Intersubunit communication and fast gate integrity are important for common gating in..., Cederholm [/bib_ref] , which would make this residue an essential part of both gating processes in ClC channels and also in the Cl − /H + ion translocation in ClC exchangers. [bib_ref] Structure of a eurkaryotic CLC transporter defines an intermediate state in the..., Feng [/bib_ref] proposed a hypothesis for the mechanism of coupled Cl − /H + transport, in which Glu ext cycles between S ext , S cen , and the extracellular environment. While occupying S cen , Glu ext interacts with Tyr cen and accepts a proton from the intracellular H + pathway. Then, following a conformational change after protonation, it would deliver the H + to the extracellular solution. Presupposing that common gating and Cl − /H + translocation are evolutionarily linked, and using the Cl − /H + transport mechanism described above as a model, [bib_ref] Molecular determinants of common gating of a ClC chloride channel, Bennetts [/bib_ref] suggested that Glu ext and Tyr cen play an important role for ClC-0 and ClC-1 common gating as they do for Cl − /H + translocation. Additionally, they proposed that conformational changes for closure of the common gating involve helices G, F, H, I, and the CBS2 domain of the adjacent subunit, resulting in an arrangement that places Glu ext (helix F) in position for hydrogen bonding with Tyr cen (helix R), locking the channel closed. In this model, helix G would function as the coordinator between protopore and subunit interface, integrating both subunits for the common gating.
In the same work, the authors reported the involvement of Tyr cen in Zn +2 inhibition and NAD + modulation of the common gate [bib_ref] Molecular determinants of common gating of a ClC chloride channel, Bennetts [/bib_ref]. This research sheds some light on the molecular determinants of the common gating of ClC channels, but much remains unclear. The pathway for the H + transport-proposed to be involved in the common gatingis not yet defined, as the suggested intracellular coordinator (Glu in ) is changed by a valine residue in ClC channels. Also, in the eukaryotic CmClC Cl − /H + exchanger, Glu in is replaced by a threonine residue that either perform this transport or this exchanger would use an alternative H + pathway [bib_ref] Structure of a eurkaryotic CLC transporter defines an intermediate state in the..., Feng [/bib_ref]. A neighboring conserved Glu residue (E291 in ClC-1), however, was proposed as a substitute to execute this function [bib_ref] Proton-coupled gating in chloride channels, Lísal [/bib_ref]. Mutation to a protonable aspartate (E291D) shifted voltage dependence to more positive values but preserved the pH dependence, whereas mutation to a neutral glutamine (E291Q) remarkably reduced voltage and pH dependence, suggesting the participation of this residue in the H + transport [bib_ref] Proton-coupled gating in chloride channels, Lísal [/bib_ref]. This assumption, however, cannot be confirmed based solely on mutagenesis experiments. The exact molecular rearrangement necessary for the common gating is another puzzle, with many parts still missing.
One recent study analyzing ClC-1/ClC-2 heterodimeric channels revealed channels with original gating properties. The common gating was abolished, with each subunit displaying individual slow gates as well as independent fast gates [bib_ref] ClC-1 and ClC-2 form hetero-dimeric channels with novel protopore functions, Stölting [/bib_ref]. These findings suggest that conformational changes underlying common gating mechanisms may originate within each protopore gate, and that fast and slow gating may in fact be linked mechanisms [bib_ref] Molecular determinants of common gating of a ClC chloride channel, Bennetts [/bib_ref] [bib_ref] ClC-1 and ClC-2 form hetero-dimeric channels with novel protopore functions, Stölting [/bib_ref]. Homodimeric channels are able to coordinate both slow gates, resulting in a single common gating, whereas heterodimeric channels lack this coordination and display individual slow gating for each subunit .
## Importance of clc channels and exchangers in cell homeostasis
ClC proteins are important for a number of physiological processes. In skeletal muscle, sodium and potassium channels provide the influx and efflux of cations necessary for propagation of the action potential, and the Cl − current generated by ClC-1 is critical for proper re-polarization of the muscle fiber [bib_ref] Inactivation of muscle chloride channel by transposon insertion in myotonic mice, Steinmeyer [/bib_ref] [bib_ref] CLC channel function and dysfunction in health and disease, Stölting [/bib_ref]. Impairment of ClC-1 function leads to myotonia, a condition characterized by delays in muscle relaxation after a contraction (Planells-Cases and [bib_ref] ClC-1 chloride channels: state-of-the-art research and future challenges, Imbrici [/bib_ref]. ClC-2 in enterocytes and ClC-Kb in the thick ascending limb of Henle's loop, working in concert with Na + -K + -ATPase, are necessary for Na + and Cl − transport from the lumen to the interstitium [bib_ref] Barttin is a cl-channel β-subunit crucial for renal cl-reabsorption and inner ear..., Estévez [/bib_ref] [bib_ref] Basolateral ClC-2 chloride channels in surface colon epithelium: regulation by a direct..., Catalán [/bib_ref]. In the inner ear, K + accumulation in the endolymph (critical for inner hair cell sensory transduction) mediated by KCNQ1/KCNE1 K + channels is only possible because ClC-K channels recycle the Cl − accumulated into the cell (by the action of Na + -K + -Cl − co-transporter) back to the interstitial fluid [bib_ref] Endocochlear potential depends on Cl-channels: mechanism underlying deafness in Bartter syndrome IV, Rickheit [/bib_ref].
In vesicular membranes of the endosomal/lysosomal pathway, Cl − /H + exchange mediated by ClCs is required for vesicular acidification, which is necessary for endocytosis, vesicle sorting and lysosomal digestion [bib_ref] Impaired acidification in early endosomes of ClC-5 deficient proximal tubule, Hara-Chikuma [/bib_ref] [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref] [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref] [bib_ref] Lysosomal pathology and osteopetrosis upon loss of H+-driven lysosomal Cl-accumulation, Weinert [/bib_ref]. In all tissues and subcellular compartments described above, and many others, ion channel and transporter activity combines to maintain homeostasis. Dysfunction in even one component of the system can lead to drastic ion imbalances that may culminate in local or systemic diseases. In the case of ClC proteins, impaired function of ClC-1, ClC-2, ClC-K, ClC-5, and ClC-7 may result in myotonia congenita, azoospermia/leukodystrophy, Bartter syndromes types 3 and 4, Dent's disease, and osteopetrosis/retinal degeneration/lysosomal storage disease, respectively.
## Mammalian clcs and human disorders
## Clc-1: a skeletal muscle chloride channel
ClC-1 was the first mammalian ClC channel identified using homology cloning from the Torpedo ClC-0 channel. ClC-1 is expressed almost exclusively in skeletal muscle [bib_ref] Primary structure and functional expression of a developmentally regulated skeletal muscle chloride..., Steinmeyer [/bib_ref] , and has the same double-barreled conformation reported for ClC-0, although with a considerably smaller conductance. Activation of the fast and slow gating of ClC-1 requires depolarization that is dependent on the Cl − and H + concentration [bib_ref] The muscle-chloride channel ClC-1 has a double-barreled appearance that is differentially affected..., Saviane [/bib_ref]. Adenosine nucleotides' inhibition of CLC-1 is regulated by oxidation and reduction. Thus, ATP inhibits only reduced ClC-1 channels by shifting the voltage-dependence of common gating to more positive potentials; this inhibition disappears upon oxidation of ClC-1 [bib_ref] ATP inhibition of CLC-1 Is controlled by oxidation and reduction, Zhang [/bib_ref]. Nucleotides bind in a putative site formed by residues from both CBS domains, and this inhibition is enhanced by low intracellular pH [bib_ref] Cytoplasmic ATP-sensing domains regulate gating of skeletal muscle ClC-1 chloride channels, Bennetts [/bib_ref] [bib_ref] Intracellular β-nicotinamide adenine dinucleotide inhibits the skeletal muscle ClC-1 chloride channel, Bennetts [/bib_ref]. This may be the mechanism by which the muscle fiber regulates channel function depending on the metabolic state. PKC and Zn +2 were also found to modulate ClC-1 function. Blocking by Zn +2 is closely related to the slow gating process. Mutation C277S locks the slow gate opened and abolishes the Zn +2 blocker effect, whereas mutation V321A reduces slow gating opening and facilitates Zn +2 blocking, suggesting that the effect of this ion is dependent on the state of the slow gating [bib_ref] Zinc inhibits human ClC-1 muscle chloride channel by interacting with its common..., Duffield [/bib_ref]. Several serine residues were identified in the C-terminal portion of ClC-1 that may mediate PKC modulation of the channel function [bib_ref] Functional study of CLC-1 mutants expressed in xenopus oocytes reveals that a..., Hsiao [/bib_ref]. PKC activators inhibit the channel, whereas PKC inhibitors increase the current, suggesting that PKC phosphorylation of the C-terminal portion is important for ClC-1 function [bib_ref] Activators of protein kinase C induce myotonia by lowering chloride conductance in..., Brinkmeier [/bib_ref] [bib_ref] Protein kinase C theta (PKCθ) modulates the ClC-1 chloride channel activity and..., Camerino [/bib_ref].
## Clc-1 and myotonia congenita
Skeletal muscle has a uniquely high resting Cl − conductance that is more than four times greater than the K + conductance [bib_ref] Muscle chloride channels, Bretag [/bib_ref]. ClC-1 is the predominant mediator of the Cl − conductance in skeletal muscle. The first insight into the physiological role of ClC-1 came from studies using myotonic goats [bib_ref] Sodium, potassium, and chloride fluxes in intercostal muscle from normal goats and..., Lipicky [/bib_ref] and myotonic adr mice [bib_ref] Inactivation of muscle chloride channel by transposon insertion in myotonic mice, Steinmeyer [/bib_ref]. Skeletal muscle fibers from these animals failed to repolarize following repeated action potentials, resulting in the so-called 'myotonic after-discharge' condition, characterized by muscle stiffness [bib_ref] On the repetitive discharge in myotonic muscle fibres, Adrian [/bib_ref]. After an action potential, Na + channels close and K + channels open to allow the ion efflux necessary for repolarization. In the T-tubules this K + accumulation (increasing of [K + ] ext ) may generate small depolarizations even after inputs from the nervous system have ceased. ClC-1 mediates Cl − conductance that prevents the K + -mediated depolarization from propagating along the sarcolemma . In myotonic fibers, the lack of ClC-1 conductance leads to autonomous fiber action potentials that keep the muscle active, delaying relaxation [bib_ref] Inactivation of muscle chloride channel by transposon insertion in myotonic mice, Steinmeyer [/bib_ref] [bib_ref] CLC channel function and dysfunction in health and disease, Stölting [/bib_ref]. Mutations in the ClC-1 gene were found in families with myotonia congenita. These mutations lead to partial or complete loss of function of ClC-1, affecting channel function in different ways depending on the mutation. A group of mutations cause a reverted voltage dependency, i.e., D136G [bib_ref] An aspartic acid residue important for voltage-dependent gating of human muscle chloride..., Fahlke [/bib_ref] , G499R [bib_ref] Mechanism of inverted activation of ClC-1 channels caused by a novel myotonia..., Zhang [/bib_ref] , C277Y [bib_ref] Disease-causing mutations C277R and C277Y modify gating of human ClC-1 chloride channels..., Weinberger [/bib_ref] , G523D [bib_ref] Electrophysiological characteristics of six mutations in hClC-1 of korean patients with myotonia..., Ha [/bib_ref]. These mutations cause the channels to activate upon hyperpolarization FIGURE 1 | Flowchart of the proposed new gating behavior of ClC-1/ClC-2 heterodimers [bib_ref] ClC-1 and ClC-2 form hetero-dimeric channels with novel protopore functions, Stölting [/bib_ref]. Homodimers present individual fast gating for each subunit and a single common gating generated by the coordination of each subunit's slow gating. In the heterodimer assembly (center), the individual protopore gating is maintained whereas coordination of each subunit's slow gating is missing. In those channels each subunit displays individual slow gating (with distinct time and voltage dependence), therefore, the common gating is not observed.
rather than deactivate like wild-type ClC-1, rendering channels with dramatically reduced or abolished currents at physiological chloride gradients. Mutation G230E [bib_ref] A mutation in autosomal dominant myotonia congenita affects pore properties of the..., Fahlke [/bib_ref] and the aforementioned C277Y alters the ion selectivity of the channel pore. The A531V has normal gating properties but has reduced expression at the plasma membrane due to an increased degradation rate [bib_ref] Myotonia congenita mutation enhances the degradation of human CLC-1 chloride channels, Lee [/bib_ref]. It was later shown by [bib_ref] The cullin 4A/B-DDB1-cereblon E3 ubiquitin ligase complex mediates the degradation of CLC-1..., Chen [/bib_ref] that a ubiquitin ligase complex (CUL4A/B-DDB1-CRBN) ubiquinates the A531V mutant leading to its subsequent degradation. To date, more than 130 mutations have been identified in the gene encoding ClC-1, and heterologous expression of mutated channels has played a valuable role in helping scientists to understand channel structure and function and disease pathogenesis [bib_ref] The non-dystrophic myotonias: molecular pathogenesis, diagnosis and treatment, Matthews [/bib_ref] [bib_ref] ClC-1 chloride channels: state-of-the-art research and future challenges, Imbrici [/bib_ref].
Myotonia congenita is the most common skeletal muscle hereditary channelopathy in humans, characterized by an atypical delay in muscle relaxation after voluntary contractions, called muscle stiffness. The myotonic stiffness is worse after rest, and improves after repetitive movements, referred as the warm-up phenomenon [bib_ref] Tonische Krämpfe in willkürlich beweglichen Muskeln in Folge von ererbter psychischer Disposition, Thomsen [/bib_ref] [bib_ref] The electrophysiology of myotonia, with a review of congenital myotonia of goats, Bryant [/bib_ref]. In humans, myotonia can be inherited as a dominant (Thomsen disease) or a recessive (Becker disease) trait, with more severe symptoms found in the latter form [bib_ref] The muscle-chloride channel ClC-1 has a double-barreled appearance that is differentially affected..., Saviane [/bib_ref]. In the dominant form, mutant subunits exert a dominant negative effect on wild-type subunits; that is, the mutant impairs (at variable levels) the function of the wild-type subunit.
Using the crystal structure of cmClC (an eukaryotic ClC exchanger) [bib_ref] Structure of a eurkaryotic CLC transporter defines an intermediate state in the..., Feng [/bib_ref] as a model for human ClC-1 allowed the identification of residues found mutated in myotonic patients in the dimer interface and in the ion conduction protopore [bib_ref] CLCN1 mutations in czech patients with myotonia congenita, in silico analysis of..., Skálová [/bib_ref]. Moreover, mutations causing the dominant-negative effect were located in or proximal to the dimer interface region. Meanwhile, mutations affecting the channel protopore do not exert the dominant-negative effect [bib_ref] CLCN1 mutations in czech patients with myotonia congenita, in silico analysis of..., Skálová [/bib_ref]. As the slow gating involves subunit interactions at the dimer interface, mutations affecting this area in only one subunit prevent the coordination necessary for the common gating and explain the impairment of the adjacent wildtype subunit. In the recessive form, both subunits are affected and ClC-1 currents may be abolished completely, leading to the more severe symptoms reported [bib_ref] The muscle-chloride channel ClC-1 has a double-barreled appearance that is differentially affected..., Saviane [/bib_ref] [bib_ref] ClC-1 chloride channels: state-of-the-art research and future challenges, Imbrici [/bib_ref].
To date, there is no specific treatment for patients with myotonia congenita. To surpass ClC-1 defect, the ideal drug should specifically enhance its Cl − currents; unfortunately, this objective seems to be far from completion [bib_ref] ClC-1 chloride channels: state-of-the-art research and future challenges, Imbrici [/bib_ref]. One early study showed that the R-isomer of CPP, a clofibric acid derivative, was able to increase Cl − conductance in voltage clamp recordings of muscle fibers [bib_ref] Opposite effects of enantiomers of clofibric acid derivative on rat skeletal muscle..., De Luca [/bib_ref]. This activity was not recognized in heterologously expressed channels, suggesting that the drug does not interact directly with ClC-1 and probably uses a muscle-specific component to exert its effect [bib_ref] Pharmacological characterization of chloride channels belonging to the ClC family by the..., Pusch [/bib_ref]. Acetazolamide (a carbonic anhydrase FIGURE 2 | ClC-1 is a major ion channel involved in the membrane resting potential of skeletal muscles. Action potentials, from motor neurons, causes the opening of L-type calcium channels (DHPR) that in turn open intracellular channels (RyR). Calcium release from both channels increases sarcoplasmic reticulum [Ca 2+ ] necessary for muscle contraction. After contraction, K + efflux repolarizes the membrane. ClC-1 chloride conductance prevents K + accumulation at the T-tubules from propagating along the sarcolemma and trigger undesirable autonomous depolarizations.
inhibitor) was also reported to be able to shift the voltagedependence of ClC-1 channel opening to more negative voltages, possibly through changes in intracellular pH, consequently enhancing Cl − conductance. However, this potentially antimyotonic effect was not effective in some mutant channels [bib_ref] Acetazolamide acts directly on the human skeletal muscle chloride channel, Eguchi [/bib_ref] [bib_ref] Functional characterization of ClC-1 mutations from patients affected by recessive myotonia congenita..., Desaphy [/bib_ref].
## Clc-2: a widely expressed clc channel
The discovery of ClC-2 came soon after ClC-1. ClC-2 is approximately 50% identical to ClC-1, and is expressed in the plasma membrane of cells from a variety of tissues, including the brain, kidney, pancreas, skeletal muscles, heart, lungs, gastrointestinal tract, and liver [bib_ref] A chloride channel widely expressed in epithelial and non-epithelial cells, Thiemann [/bib_ref]. ClC-2 opens in a very short time course upon hyperpolarization. Its voltagedependent gating is modulated by the concentration of Cl − and H + . Increase in the intracellular concentration of Cl − shifts the voltage-dependence to a more positive voltage, activating the channel. ClC-2 is also activated by mild decreases in extracellular pH, although a further decrease in pH reduces current .
ClC-2 can bind to the accessory molecule GlialCAM, an adhesion molecule, in several glial cell types. This interaction is not required for the channel to function, but rather modifies the channel gating properties [bib_ref] GlialCAM, a protein defective in a leukodystrophy, serves as a ClC-2 Cl-channel..., Jeworutzki [/bib_ref]. GlialCAM also binds to Mlc1 (a membrane protein involved in megalencephalic leukoencephalopathy with subcortical cysts, a type of leukodystrophy) and docks both complexes (GlialCAM-ClC-2 and GlialCAM-Mlc1) at cell-cell junctions [bib_ref] Mutant GlialCAM causes megalencephalic leukoencephalopathy with subcortical cysts, benign familial macrocephaly, and..., López-Hernández [/bib_ref] [bib_ref] GlialCAM, a protein defective in a leukodystrophy, serves as a ClC-2 Cl-channel..., Jeworutzki [/bib_ref] [bib_ref] Disrupting MLC1 and GlialCAM and ClC-2 interactions in leukodystrophy entails glial chloride..., Hoegg-Beiler [/bib_ref]. Co-expression of GlialCAM with ClC-2 increases currents and almost eliminates the inward rectification, rendering ClC-2 channels nearly constitutively open [bib_ref] GlialCAM, a protein defective in a leukodystrophy, serves as a ClC-2 Cl-channel..., Jeworutzki [/bib_ref]. Disruption of either GlialCAM or Mlc1 affects the expression and localization of ClC-2 (Hoegg-Beiler et al., 2014). Mutations of either GlialCAM or Mcl1 genes lead to megalencephalic leukoencephalopathy, a type of leukodystrophy characterized by early-onset macrocephaly and delayed-onset neurologic deterioration, symptoms comparable to the neurological phenotype of ClC-2 disruption [bib_ref] Mutant GlialCAM causes megalencephalic leukoencephalopathy with subcortical cysts, benign familial macrocephaly, and..., López-Hernández [/bib_ref].
ClC-2 gating is affected by ATP and, like ClC-1, ATP changes the voltage dependence of the common gating. Whole-cell patch clamp recordings show slow activation and deactivation times. Single channel recordings exhibit longer periods of closed states of the common gating when high levels of intracellular ATP are present. This effect, however, does not change the open probability of the channel.
## Clc-2 in azoospermia and leukodystrophy
In the testes, tight junctions between Sertoli cells isolate the adluminal compartment of the seminiferous tubules from the blood irrigation (blood-testis barrier). Because of this barrier, maturation and differentiation of spermatogonia into sperm cells require a close physical contact with Sertoli cells that are also responsible for the nourishment of the germ cells during this process. Disruption of ClC-2 function results in transepithelial transport defect in Sertoli cells and subsequent degeneration of male germ cells (azoospermia), as observed in ClC-2 knock-out (KO) mice [bib_ref] Chloride channelopathies of ClC-2, Bi [/bib_ref].
ClC-2 KO mice also develop leukodystrophy (the general term for diseases affecting the growth or maintenance of the white matter), which culminates with gradual development of vacuoles in the myelin sheath of the central nervous system, worsening with age [bib_ref] Leukoencephalopathy upon disruption of the chloride channel ClC-2, Blanz [/bib_ref]. Human patients carrying mutations that disrupt ClC-2 function develop similar leukodystrophy symptoms [bib_ref] Brain white matter oedema due to ClC-2 chloride channel deficiency: an observational..., Depienne [/bib_ref] ; in one patient, azoospermia was found together with a subclinical leukodystrophy (Di [bib_ref] Subclinical leukodystrophy and infertility in a man with a novel homozygous CLCN2..., Bella [/bib_ref]. This was the first case report demonstrating azoospermia and leukodystrophy in a patient with ClC-2 mutation.
## Other controversial physiological roles
ClC-2 is also expressed in epithelial cells of the gastrointestinal tract and lungs. In the past, ClC-2 was proposed to play a role in Cl − efflux at the apical membrane of epithelial cells of these tissues, working as an alternative pathway to CFTR-dependent Cl − secretion. However, the intestinal phenotype observed in CFTR-KO mice was not aggravated in double KO mice, in the absence of both CFTR and ClC-2. Instead, double KO mice survived better than CFTR-KO mice [bib_ref] Additional disruption of the ClC-2 Cl(-) channel does not exacerbate the cystic..., Zdebik [/bib_ref]. Later on, it was demonstrated that ClC-2 localizes at the basolateral membrane of enterocytes, facilitating water and salt absorption [bib_ref] Basolateral ClC-2 chloride channels in surface colon epithelium: regulation by a direct..., Catalán [/bib_ref]. In the basolateral membrane, ClC-2 is proposed to move Cl − in the opposite direction of CFTR, e.g., moving Cl − from the cell to the interstitium. Loss of ClC-2 in CFTR-KO mice would then increase Cl − concentration inside the cell, facilitating Cl − efflux in the apical compartment by an alternative pathway and compensating for the loss of CFTR from the apical membrane. These and other reports [bib_ref] Basolateral ClC-2 chloride channels in surface colon epithelium: regulation by a direct..., Catalán [/bib_ref] [bib_ref] Basolateral localization of native ClC-2 chloride channels in absorptive intestinal epithelial cells..., Peña-Münzenmayer [/bib_ref] provide convincing data for the basolateral localization of ClC-2 in intestinal epithelia. ClC-2 could play the same role in the FIGURE 3 | ClC-2 aids in water absorption in intestinal epithelia. In colonic enterocytes, chloride absorbed from the intestinal lumen (via SLC26A3 chloride/bicarbonate exchanger) is transported to the interstitium through ClC-2. Sodium enters the cell via ENaC channels or sodium/proton exchangers and is transported to the interstitium through the Na+/K+ ATPase. High NaCl gradient at the interstitium induces osmotic water absorption from the lumen. lung epithelium, although its precise localization is still not conclusive.
ClC-2 is also expressed in neurons and glial cells, where it is proposed to lower the intracellular concentration of Cl − . ClC-2 would be activated after a Cl − influx mediated by hyperpolarizing GABA currents. ClC-2, then, would extrude the excess of intracellular Cl − down to its electrochemical equilibrium helping in the maintenance of a Cl − gradient favorable to cell hyperpolarization by GABA currents [bib_ref] Alteration of GABA A receptor function following gene transfer of the CLC-2..., Staley [/bib_ref] [bib_ref] Regulation of fastspiking basket cell synapses by the chloride channel ClC-2, Földy [/bib_ref] [bib_ref] ClC-2 voltage-gated channels constitute part of the background conductance and assist chloride..., Rinke [/bib_ref]. This theory, however, was questioned by a study using a computational model-based on ClC-2 parameters previously characterized in CA1 pyramidal cells-simulating physiological conditions which showed ClC-2 actually mediating chloride influx, directly reducing cell excitability [bib_ref] ClC-2 channels regulate neuronal excitability, not intracellular chloride levels, Ratté [/bib_ref].
The retinal pigment epithelia (RPE) are responsible for forming the blood-organ barrier in the eye, creating the optimal microenvironment for photoreceptor function. Loss of retinal photoreceptors induces retinal degeneration. Loss of ClC-2 function has been proposed to affect transepithelial transport in the RPE by disrupting microenvironment ion homeostasis, resulting in photoreceptor degeneration [bib_ref] Chloride channelopathies of ClC-2, Bi [/bib_ref]. Studies on ClC-2 KO mice revealed retinal degeneration, indicating an important role for this channel in RPE. This degenerative phenotype suggests the disruption of ion homeostasis in this tissue .
Previously, several other functions were thought to be assigned to ClC-2. Suggested roles in gastric acid secretion [bib_ref] Localization of ClC-2 cl-channels in rabbit gastric mucosa, Sherry [/bib_ref] and lung development [bib_ref] CIC-2: a developmentally dependent chloride channel expressed in the fetal lung and..., Murray [/bib_ref] were supported neither by experimental data nor by ClC-2 KO mice phenotype. A role in epilepsy was also considered, but after the retraction of a widely cited paper correlating ClC-2 mutations to idiopathic generalized epilepsy, there is no credible evidence for a ClC-2 role in human epilepsy. This is consistent with the lack of seizures observed in ClC-2 KO mice [bib_ref] Leukoencephalopathy upon disruption of the chloride channel ClC-2, Blanz [/bib_ref].
## Clc-ka and clc-kb: largely open clc channels that require a β-subunit
ClC-Ka and ClC-Kb (-K1 and -K2 in rodents) are two closely related ClC channels (around 90% identical) [bib_ref] Two isoforms of a chloride channel predominantly expressed in thick ascending limb..., Adachi [/bib_ref] [bib_ref] Two highly homologous members of the ClC chloride channel family in both..., Kieferle [/bib_ref] , expressed almost entirely in nephrons and in the stria vascularis of the inner ear [bib_ref] Barttin is a cl-channel β-subunit crucial for renal cl-reabsorption and inner ear..., Estévez [/bib_ref]. Different from the other mammalian ClC channels, the two ClC-K isoforms lack the 'gating glutamate, ' displaying halide selectively sequence of Br − > Cl − > I − [bib_ref] Two isoforms of a chloride channel predominantly expressed in thick ascending limb..., Adachi [/bib_ref]. ClC-K channels have only a slight voltagedependent gating and hence are open over a broad voltage range [bib_ref] Barttin is a cl-channel β-subunit crucial for renal cl-reabsorption and inner ear..., Estévez [/bib_ref] [bib_ref] Cell biology and physiology of CLC chloride channels and transporters, Stauber [/bib_ref]. The first heterologously expressed ClC-Ka and -Kb channels (and also the mice ClC-K2) failed to display any conductance [bib_ref] Two highly homologous members of the ClC chloride channel family in both..., Kieferle [/bib_ref] , which raises questions about the necessity of a β-subunit, given that immunohistochemistry [bib_ref] Localization and induction by dehydration of ClC-K chloride channels in the rat..., Vandewalle [/bib_ref] and disease-causing mutations [bib_ref] Mutations in the chloride channel gene, CLCNKB, cause bartter's syndrome type III, Simon [/bib_ref] clearly indicate their participation in transepithelial salt transport.
Barttin, a 40 kDa and 320-residue protein containing two transmembrane domains and a long intracellular C-terminal domain was identified as the required accessory protein for human ClC-K proteins to be functional [bib_ref] Barttin is a cl-channel β-subunit crucial for renal cl-reabsorption and inner ear..., Estévez [/bib_ref]. Barttin is essential for channel function, stability, and trafficking to the correct membrane area within the cell [bib_ref] Barttin is a cl-channel β-subunit crucial for renal cl-reabsorption and inner ear..., Estévez [/bib_ref] [bib_ref] Barttin increases surface expression and changes current properties of ClC-K channels, Waldegger [/bib_ref] [bib_ref] Barttin modulates trafficking and function of ClC-K channels, Scholl [/bib_ref]. The transmembrane region of barttin is important for its association with ClC-K proteins as well as trafficking to the plasma membrane, whilst the initial part of the C-terminal domain is essential for channel conductance activation [bib_ref] Barttin modulates trafficking and function of ClC-K channels, Scholl [/bib_ref].
To date, there has been little investigation of interactions between barttin and ClC-K proteins, but two helices of ClC-K are proposed to interact with the transmembrane domain of barttin [bib_ref] Barttin binds to the outer lateral surface of the ClC-K2 chloride channel, Tajima [/bib_ref]. To further investigate the molecular determinants of barttin/ClC-K interactions, [bib_ref] Tryptophan scanning mutagenesis identifies the molecular determinants of distinct barttin functions, Wojciechowski [/bib_ref] used tryptophan scanning mutagenesis to identify amino acids in the transmembrane domains of barttin essential for ClC-K function. Taking into account only normally expressing barttin mutants (some mutants were misfolded or had low expression), substitution of six amino acids (three in each of the transmembrane domains) affected ClC-K/barttin trafficking to the membrane. In contrast, several mutations directly affected ClC-K function. ClC-K currents were abolished when coexpressed with 12 barttin mutants (nine at the first and three at the second transmembrane domains) while two tryptophan insertions at the second transmembrane caused reduced current amplitudes. As most inactivating mutants had tryptophan insertions at the first transmembrane domain, the authors suggest that this domain is critical for activation of ClC-K channels [bib_ref] Tryptophan scanning mutagenesis identifies the molecular determinants of distinct barttin functions, Wojciechowski [/bib_ref].
Co-expressed ClC-K/barttin channels display a very high Cl − conductance (∼40 pS compared to ∼1 pS for ClC-1 and ∼3 pS for ClC-2), modulated by extracellular pH and Ca +2 concentration; function is inhibited by H + and activated by Ca +2 . However, the physiological importance of these modulations are still unclear. ClC-K/barttin localizes at the basolateral membranes of both the thin and thick ascending limbs of Henle's loop, and in marginal cells of the stria vascularis of the inner ear [bib_ref] Barttin is a cl-channel β-subunit crucial for renal cl-reabsorption and inner ear..., Estévez [/bib_ref]. ClC-K1 was also found in the apical membrane of the thin ascending limb of Henle's loop .
## Clc-k in renal salt loss and deafness
ClC-Kb/barttin is mainly expressed in basolateral membranes of the thick ascending limb of Henle's loop, where it is involved in the reabsorption of salt and, consequently, water . In this part of the nephron, the Na + electrochemical gradient (created by basolateral Na + /K + pump) drives the secondary active transport of NKCC2 (present in the apical membrane), accumulating Na + , Cl − , and K + into the cell. K + is extruded back to the lumen through ROMK K + channels (also present in the apical membrane), whereas Na + and Cl − are reabsorbed by the interstitial fluid through the Na + /K + pump and ClC-Kb channels, respectively. Thus, the end product of this system is the reabsorption of NaCl into the blood stream .
In the inner ear, both ClC-K isomers are expressed in the basolateral membrane of marginal cells of the stria vascularis. This multilayered epithelium is responsible for both the high concentration of K + and the positive potential (about 100 mV higher than normal extracellular fluids) of the endolymph of the scala media, both of which are important properties for hearing. In marginal cells-the more apical layer in the stria vascularis-Na + /K + pumps and NKCC1 transporters build up K + and Cl − inside the cells. ClC-K/barttin channels recycle Cl − back to the interstitial fluid, while apical KCNQ1/KCNE1 K + channels secrete the excess of potassium ions into the endolymph [bib_ref] Endocochlear potential depends on Cl-channels: mechanism underlying deafness in Bartter syndrome IV, Rickheit [/bib_ref].
In agreement with the transport models involving ClC-K/barttin channels, mutations in the gene encoding ClC-Kb cause salt-losing Bartter syndrome type III [bib_ref] Mutations in the chloride channel gene, CLCNKB, cause bartter's syndrome type III, Simon [/bib_ref] , characterized by hypokalemia, metabolic alkalosis and secondary hyperaldosteronism with normal or low blood pressure [bib_ref] ClC-K chloride channels: emerging pathophysiology of bartter syndrome type 3, Andrini [/bib_ref]. Mutations in the gene encoding barttin cause Bartter syndrome type IV that combines the salt waste with congenital deafness, since both ClC-K proteins are non-functional in the FIGURE 4 | ClC-K channels are expressed in kidney and inner ear. (A) At the nephrons, luminal NKCC2 transporters build up Na + , K + and Clinto the cells. K + flows back to the lumen through ROMK1 channels; Na + and Clare reabsorbed to the bloodstream separately through Na+/K+ ATPase and ClC-Kb channels, respectively. (B) In the Stria Vascularis, Na + , K + and Clare transported into the cells by basolateral NKCC1 transporters. Na + and Clare recycled back to the interstitium by Na+/K+ ATPase and both ClC-Ks isomers, respectively. K + flows through KCNQ1/KCNE1 channels and accumulates into the endolymph, a condition required for sensory transduction in inner hair cells.
absence of barttin . When disruption occurs in only one of the ClC-K channels, as it does in ClC-Kb mutations in Bartter type III, hearing is preserved; the other isomer channel still provides the necessary Cl − recycling. Deafness occurs only on disruption of both ClC-K channels or upon disruption of barttin [bib_ref] Salt wasting and deafness resulting from mutations in two chloride channels, Schlingmann [/bib_ref].
Although there are no reports of patients presenting mutations only in ClC-Ka, two patients presenting symptoms similar to those accompanying Bartter syndrome type IV-severe renal salt loss and sensorineural deafness-were described with loss-of-function mutations in both ClC-K isomers instead of barttin [bib_ref] Salt wasting and deafness resulting from mutations in two chloride channels, Schlingmann [/bib_ref] [bib_ref] Molecular analysis of digenic inheritance in Bartter syndrome with sensorineural deafness, Nozu [/bib_ref].
## Clc-k involvement in cardiovascular diseases
Polymorphisms in ClC-Ka and -Kb genes were described, and their relationship with cardiovascular diseases was analyzed. ClC-Kb gene polymorphism T481S increases currents in heterologously expressed channels by approximately 20-fold . This may lead to increased salt reabsorption in the thick ascending limb of Henle's loop, suggesting a possible connection with hypertension. However, several cohort studies found discrepant results, and a link between this activating polymorphism and hypertension is still lacking [bib_ref] No association with hypertension of CLCNKB and TNFRSF1B polymorphisms at a hypertension..., Speirs [/bib_ref] [bib_ref] The functional variant of the CLC-Kb channel T481S is not associated with..., Fava [/bib_ref] [bib_ref] CLCNKB-T481S and essential hypertension in a ghanaian population, Sile [/bib_ref]. One frequent polymorphism in the ClC-Ka gene (R83G) was linked to heart failure. R83G was reported to reduce ClC-Ka currents by about 50%, and was statistically associated with heart failure in three independent Caucasian cohorts [bib_ref] Loss-of-function DNA sequence variant in the CLCNKA chloride channel implicates the cardio-renal..., Cappola [/bib_ref]. However, a functional link between this half-loss-of-function polymorphism and heart failure is still not established.
ClC-K/barttin channels are promising candidates for therapeutic drugs. As ClC-Kb is involved in salt and water reabsorption in the thick ascending limb of Henle's loop, drugs blocking ClC-Kb/barttin function could reduce renal salt and water reabsorption, which would decrease blood volume and consequently reduce blood pressure. In the inner ear, hearing depends on the depolarization of mechanosensitive hair cells. Different from other excitable cells that use Na + currents for depolarization, depolarization of hair cells is mediated by K + influx.
Drugs capable of increasing ClC-K/barttin function in the stria vascularis would increase endolymph K + concentration, and therefore could be used to treat hearing disorders. However, due to expression of ClC-K/barttin channels in both the kidney and inner ear, it will be difficult to develop specific drugs without undesirable side effects. Recently, while testing new benzofuran derivatives designed to block ClC-K function, [bib_ref] Kidney CLC-K chloride channels inhibitors: structure-based studies and efficacy in hypertension and..., Liantonio [/bib_ref] described the most potent and selective ClC-K blocker discovered to date (SRA-36). This compound is able to inhibit not only wild-type channels, but also the Cl − currents of polymorphic ClC-K channels associated with hypertension [bib_ref] Kidney CLC-K chloride channels inhibitors: structure-based studies and efficacy in hypertension and..., Liantonio [/bib_ref]. Although several studies have made significant progress on the identification of compounds modulating ClC-K channel function [bib_ref] Activation and inhibition of kidney CLC-K chloride channels by fenamates, Liantonio [/bib_ref] [bib_ref] Kidney CLC-K chloride channels inhibitors: structure-based studies and efficacy in hypertension and..., Liantonio [/bib_ref] [bib_ref] Molecular determinants of differential pore blocking of kidney CLC-K chloride channels, Picollo [/bib_ref] , there are not yet therapeutic drugs available.
## Clc-5: a clc exchanger of early endosomes
ClC-5 is the most well-studied member of the second branch of the ClC family. It was identified independently by linkage analysis of patients with Dent's disease [bib_ref] Isolation and partial characterization of a chloride channel gene which is expressed..., Fisher [/bib_ref] and by homology cloning [bib_ref] Cloning and functional expression of rat CLC-5, a chloride channel related to..., Steinmeyer [/bib_ref]. Unlike ClC-3 and ClC-4, ClC-5 has a more restricted tissue distribution, localizing mostly in renal and intestinal epithelia [bib_ref] Cloning and functional expression of rat CLC-5, a chloride channel related to..., Steinmeyer [/bib_ref] [bib_ref] Tissue distribution and subcellular localization of the ClC-5 chloride channel in rat..., Vandewalle [/bib_ref]. In the kidney, ClC-5 is mostly expressed in acid-transporting intercalated cells in distal nephron and in PTCs. In PTCs, ClC-5 is co-localized with V-type H + -ATPase at early and recycling endosomes, with only a small amount found at the surface membrane of brush cells. In intestinal epithelia, ClC-5 also co-localizes with the proton ATPase in apical endosomes [bib_ref] Tissue distribution and subcellular localization of the ClC-5 chloride channel in rat..., Vandewalle [/bib_ref].
Although it is mainly localized in apical endosomes, a reasonable amount of ClC-5 can be found at the cell surface upon heterologous overexpression, where it can be biophysically analyzed [bib_ref] Cloning and functional expression of rat CLC-5, a chloride channel related to..., Steinmeyer [/bib_ref] [bib_ref] Mutational analysis demonstrates that ClC-4 and ClC-5 directly mediate plasma membrane currents, Friedrich [/bib_ref]. ClC-5 is a 2Cl − /1H + exchanger [bib_ref] Chloride/ proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5, Picollo [/bib_ref] [bib_ref] Voltage-dependent electrogenic chloride/ proton exchange by endosomal CLC proteins, Scheel [/bib_ref] with an anion conductance sequence of Cl − > Br − > I − , and displays strong outwardly rectifying current that is decreased by low extracellular pH, similar to the closely related ClC-4 [bib_ref] Cloning and functional expression of rat CLC-5, a chloride channel related to..., Steinmeyer [/bib_ref] [bib_ref] Mutational analysis demonstrates that ClC-4 and ClC-5 directly mediate plasma membrane currents, Friedrich [/bib_ref]. Similar to other ClC exchangers, neutralization of the 'gating glutamate' results in uncoupled Cl − passive conductance and eliminates voltage dependence [bib_ref] Chloride/ proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5, Picollo [/bib_ref] [bib_ref] Voltage-dependent electrogenic chloride/ proton exchange by endosomal CLC proteins, Scheel [/bib_ref]. Interestingly, however, the neutralization of Glu int -the putative intracellular proton acceptor-results in disruption of both Cl − and H + transport in ClC-5, creating a transport-deficient protein.
That behavior differs from the prokaryotic exchanger ecClC, which exhibits Cl − passive conductance upon either 'gating glutamate' or 'proton glutamate' mutations . [bib_ref] Glutamate 268 regulates transport probability of the anion/proton exchanger ClC-5, Grieschat [/bib_ref] found similar results by neutralizing either Glu ext or Glu int .
When ClC-5 is expressed heterologously, replacement of extracellular Cl − with SCN − leads to uncoupling of anion transport but does not affect proton transport. The change from extracellular Cl − to SCN − led to increased current amplitudes, and this effect was ∼4-fold higher when intracellular pH was reduced. The effect of intracellular protons was suggested to be related to the protonation of Glu ext . Neutralizing either Glu ext or Glu int (E211C; E268C) eliminated the effect of low pH. With neutralization of Glu ext the cysteine side chains are not available for protonation. With neutralization of Glu int cysteine is unable to supply protons to Glu ext [bib_ref] Glutamate 268 regulates transport probability of the anion/proton exchanger ClC-5, Grieschat [/bib_ref]. In the case of the E268C mutant, transport was restored to wildtype levels after reaction of a negatively charged and protonable MTSES compound with C268, indicating that the ability of Glu int to protonate Glu ext regulates transport probability in ClC-5 [bib_ref] Glutamate 268 regulates transport probability of the anion/proton exchanger ClC-5, Grieschat [/bib_ref].
The cytoplasmic domain of ClC-5 was found to bind adenosine nucleotides in a site located between the CBS domains. As binding of AMP, ADP or ATP occurs with similar affinities, the physiological role of nucleotide binding remains unclear [bib_ref] Nucleotide recognition by the cytoplasmic domain of the human chloride transporter ClC-5, Meyer [/bib_ref]. Also, in the region between its two CBS domains, ClC-5 carries a PY-motif known to bind WW-domains of ubiquitin ligases [bib_ref] An internalization signal in ClC-5, an endosomal Cl-channel mutated in Dent's disease, Schwake [/bib_ref]. Point mutations that eliminate the PY-motif of ClC-5 double the currents and increase cell surface localization upon heterologous expression [bib_ref] An internalization signal in ClC-5, an endosomal Cl-channel mutated in Dent's disease, Schwake [/bib_ref]. However, knock-in mice with a point mutation disrupting the PY-motif lack any of the effects observed in vitro .
## Clc-5 and dent's disease
Dent's disease is a rare X-linked kidney disorder associated with low molecular weight proteinuria, hyperphosphaturia, hypercalciuria, kidney stones, and nephrocalcinosis . After the identification of ClC-5 mutations as the cause of Dent's disease [bib_ref] Characterisation of renal chloride channel, CLCN5, mutations in hypercalciuric nephrolithiasis (kidney stones)..., Lloyd [/bib_ref] , more than 100 such mutations were described [bib_ref] ClC-5: physiological role and biophysical mechanisms, Pusch [/bib_ref]. Most mutations in ClC-5 are missense and non-sense mutations, with many of them located at or near the subunit's interface, resulting in non-functional truncated proteins [bib_ref] Modeling study of human renal chloride channel (hCLC-5) mutations suggests a structural-functional..., Wu [/bib_ref] [bib_ref] Cell biology and physiology of CLC chloride channels and transporters, Stauber [/bib_ref]. Two missense mutations (G212A and E267A) were analyzed regarding their functional consequences. The particular interest in these mutations is explained by their close proximity to Glu ext (E211) and Glu int (E268). Both mutations result in impaired endosomal acidification, however, the causes are distinct. For the G212A mutant, a shift to more depolarizing potentials is the cause of reduced transport, whereas in E267A mutant the inability to transport intracellular protons results in an incomplete transport cycle [bib_ref] Mutations associated with dent's disease affect gating and voltage dependence of the..., Alekov [/bib_ref].
Proximal tubule cells are the main site for re-uptake of low molecular weight proteins from the primary urine filtrated at the glomeruli. The co-localization of ClC-5 and H + -ATPase in PTCs, and the loss of low molecular weight proteins in Dent's disease patients, suggests that ClC-5 might be involved in early tubular endocytosis in nephrons. To better investigate this hypothesis, two ClC-5 KO mice models were independently generated. Both models showed loss of low molecular weight proteins and high levels of retinal-and Vitamin D-binding proteins in the urineproteins also elevated in the urine of Dent's disease patients , and defective endocytosis in the proximal tubule [bib_ref] Mice lacking renal chloride channel, CLC-5, are a model for Dent's disease,..., Wang [/bib_ref]. Interestingly, only one of the mouse models displayed hypercalciuria and interstitial calcium accumulation [bib_ref] Mice lacking renal chloride channel, CLC-5, are a model for Dent's disease,..., Wang [/bib_ref]. In vivo endocytosis experiments showed that both fluid-phase and receptor-mediated endocytosis were severely reduced, and that the apical expression of NHE3 (Na + -H + exchanger) and NaPi-2a (coupled Na + -Pi cotransporter) were also reduced, all in a cell-autonomous effect of ClC-5 disruption.
Moreover, the protein megalin-an endocytotic receptor responsible for the endocytosis of several proteins and other substances-and its co-receptor cubilin were also decreased in the brush border membrane of PTCs from ClC-5 KO mice FIGURE 5 | Proposed localization of intracellular CLC exchangers to the endosomal/lysosomal pathway. ClC-5 localizes to earlier compartments of the pathway; ClC-3 and ClC-6 localize to later endosome compartments; ClC-7/Ostm1 localizes to lysosomes, the most acidic compartment. ClC-4 localization is still unclear. The ATP-proton pump (pink) acidifies the compartments, increasing protons concentration down the pathway. ClC exchangers (green) provide the shunt current in early endosomal compartments and accumulate chloride in lysosomes. [bib_ref] Loss of chloride channel ClC-5 impairs endocytosis by defective trafficking of megalin..., Christensen [/bib_ref]. Although almost all Dent's disease patients present low molecular weight proteinuria, with values ranging from 100-to over 1000-fold of the normal excretion values, the other clinical features show great variability [bib_ref] X-linked hypercalciuric nephrolithiasis: clinical syndromes and chloride channel mutations, Scheinman [/bib_ref] [bib_ref] Dent's disease: clinical features and molecular basis, Claverie-Martín [/bib_ref].
## Role of clc-5 in low weight proteinuria, hyperphosphaturia, and hypercalciuria
Luminal acidification is necessary for proper endosome function [bib_ref] Vacuolar ATPase activity is required for endosomal carrier vesicle formation, Clague [/bib_ref] and ClC-5 was thought to provide the electrical shunt necessary for the acidification of endosomes by proton pumps. Cl − ions transported by ClC-5 provide the negative charge necessary for neutralization of protons accumulating at the lumen of endosomes by the proton pump (the electrical shunt), thus maintaining the acidification process [bib_ref] Impaired acidification in early endosomes of ClC-5 deficient proximal tubule, Hara-Chikuma [/bib_ref]. Indeed, ATP-induced acidification in endosomes from ClC-5 KO animals was reduced compared to wild-type animals [bib_ref] Endosomal chloride-proton exchange rather than chloride conductance is crucial for renal endocytosis, Novarino [/bib_ref]. Furthermore, endocytosis experiments with cultured PTCs using fluorescent-tagged markers for early/recycling and late endosomes showed a reduction in both acidification and Cl − accumulation in early, but not in late, endosomes [bib_ref] Impaired acidification in early endosomes of ClC-5 deficient proximal tubule, Hara-Chikuma [/bib_ref]. A defect in endocytosis was also observed in cultured PTCs from ClC-5 KO mice .
Processes underlying the other symptoms of Dent's diseasesuch as hyperphosphaturia, hypercalciuria, and kidney stones-are more complex. Reduced megalin expression at the brush border membrane of PTCs due to ClC-5 disruption impairs the endocytosis of PTH. Accumulation of PTH at the renal tubules stimulates PTH receptors, which in turn results in degradation and internalization of NaPi-2a transporters, causing a reduction of phosphate re-absorption. Therefore, hypophosphatemia/hyperphosphaturia is observed in both Dent's disease patients [bib_ref] Dent's disease: clinical features and molecular basis, Claverie-Martín [/bib_ref] and ClC-5 KO mice. About 30% of Dent's disease patients display hypophosphatemia due to loss of phosphate in the urine [bib_ref] Dent's disease: clinical features and molecular basis, Claverie-Martín [/bib_ref].
Proximal tubule cells are also the main site for vitamin D metabolism, which plays a critical role in Ca 2+ homeostasis. In these cells, the inactive precursor 25(OH)-VitD 3 is converted to the active form 1,25(OH) 2 -VitD 3 by the mitochondrial enzyme 1α-hydroxylase, which is stimulated by PTH [bib_ref] Positive and negative regulations of the renal 25-hydroxyvitamin D3 1alpha-hydroxylase gene by..., Murayama [/bib_ref]. Megalin mediates the endocytosis of both active and inactive forms of Vitamin D. Thus, in ClC-5 KO mice, whose lack of ClC-5 results in low megalin expression in the brush border of PTCs and impaired endocytosis, two stimuli may upregulate 1α-hydroxylase expression: (1) overstimulation of PTH receptors, and (2) decreased endocytosis of the active form 1,25(OH) 2 -VitD 3 , as this form represses enzyme transcription [bib_ref] Positive and negative regulations of the renal 25-hydroxyvitamin D3 1alpha-hydroxylase gene by..., Murayama [/bib_ref] [bib_ref] Kidney-specific upregulation of vitamin D3 target genes in ClC-5 KO mice, Maritzen [/bib_ref]. Meanwhile, reduced endocytosis of the inactive form 25(OH)-VitD 3 is also in place, which would cause downregulation of 1α-hydroxylase expression.
As regulation of 1,25(OH) 2 -VitD 3 levels is governed by these two opposing mechanisms, it was hypothesized that the balance between precursor levels and those of its converting enzyme will determine the presence-or not-of hypercalciuria . If higher levels of 1α-hydroxylase lead to higher levels of 1,25(OH) 2 -VitD 3 in the serum, more calcium will be absorbed in the intestine; therefore, more calcium will be excreted in the urine, resulting in hypercalciuria and kidney stones. Indeed, patients with Dent's disease display a high prevalence (∼90%) of hypercalciuria [bib_ref] Dent's disease: clinical features and molecular basis, Claverie-Martín [/bib_ref] , as well as elevated levels of 1,25(OH) 2 -VitD 3 [bib_ref] X-linked hypercalciuric nephrolithiasis: clinical syndromes and chloride channel mutations, Scheinman [/bib_ref].
Moreover, ClC-5 KO mouse models from [bib_ref] Mice lacking renal chloride channel, CLC-5, are a model for Dent's disease,..., Wang [/bib_ref] presenting hypercalciuria and renal calcium deposits also displayed high levels of serum 1,25(OH) 2 -VitD 3 [bib_ref] Mice lacking renal chloride channel, CLC-5, are a model for Dent's disease,..., Wang [/bib_ref]. In contrast,did not identify hypercalciuria in their ClC-5 KO mice, which displayed reduced levels of serum 1,25(OH) 2 -VitD3 [bib_ref] Kidney-specific upregulation of vitamin D3 target genes in ClC-5 KO mice, Maritzen [/bib_ref]. There is a lack of reports associating hypercalciuria and calcium deposits with serum levels of Vitamin D. Prospective studies in large cohorts would be a valuable tool in the search for pathophysiological mechanisms underlying Dent's disease. Dent's disease is an excellent example of how a primary defect (impaired endocytosis) in a restricted group of cells (PTCs) can lead to a cascade of serious secondary complications (phosphaturia, calciuria, kidney stones, and nephrocalcinosis).
## Clc-5 as a cl − /h + exchanger: new insights on its role in endosomes
ClC proteins are involved in the acidification of both early and late endosomes. Endosomal acidification is a process mediated by active proton influx carried by the H + -ATPase. The inward H + movement requires a charge balance, which can be achieved both by outward movement of cations such as K + , and by inward movement of anions such as Cl − . Extensive experimental data suggest that Cl − is the principal ion providing the electrical shunt for luminal acidification of endosomes [bib_ref] Protein kinase A regulates chloride conductance in endocytic vesicles from proximal tubule, Bae [/bib_ref] [bib_ref] Chloride channels of intracellular organelles, Al-Awqati [/bib_ref] [bib_ref] Regulation of organelle acidity, Grabe [/bib_ref]. After ClC-5 was determined to be a Cl − /H + exchanger and not a Cl − channel [bib_ref] Chloride/ proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5, Picollo [/bib_ref] [bib_ref] Voltage-dependent electrogenic chloride/ proton exchange by endosomal CLC proteins, Scheel [/bib_ref] , its role as an electrical shunt for proton pumps was questioned; such an exchanger would provide a counter-current of H + , opposing the ATP-driving accumulation of protons.
To assess the consequences of this new CIC-5 feature, knockin mice were generated carrying a point mutation in the 'gating glutamate' (E211A) that uncouples Cl − and H + transport, converting ClC-5 to a pure passive Cl − conductor (called ClC-5 unc ; for uncoupled) . Surprisingly, these mice presented normal renal endosomal acidification, but also an impaired proximal tubular endocytosis similar to that found in ClC-5 KO mice. Two facts suggest that other parameters such as Cl − concentration may play a critical role in endocytosis. First, ClC-5 unc mice presented phenotypes similar to the ClC-5 KO group, including hypercalciuria and hyperphosphaturia. Second, PTC endosomes from ClC-5 unc showed normal acidification but impaired endocytosis. Recently, a patient with Dent's disease was identified as carrying a similar mutation in the 'gating glutamate' (E211Q) [bib_ref] Japanese dent disease has a wider clinical spectrum than dent disease in..., Sekine [/bib_ref]. Further support for the role of Cl − concentration in endocytosis comes from mathematical models of simplified vesicles (containing a proton pump, a proton leak and either a Cl − /H + exchanger or a Cl − channel) predicting that coupled transport would provide a FIGURE 6 | Simplified model of the effect of impaired endocytosis due to ClC-5 dysfunction in proximal tubular cells. In the early proximal tubule, PTH and vitamin D3 (both precursor and active molecules) are filtered into the primary urine. Those molecules are generally reabsorbed by megalin-mediated endocytosis and subsequently degraded in lysosomes. The megalin-mediated pathway is severely impaired in ClC-5 dysfunction. Under normal conditions 25(OH)-vitamin D 3 is metabolized by the mitochondrial enzyme 1α-hydroxylase to the active hormone 1,25(OH) 2 -vitaminD 3 . In the late proximal tubules, accumulation of PTH (due to impaired megalin-mediated endocytosis) overstimulates PTH receptors, which stimulate internalization and degradation of NaPi-2a transporters (green plus), reducing phosphate re-absorption-hyperphosphaturia. Overstimulated PTH receptors also upregulate the transcription of the mitochondrial enzyme 1α-hydroxylase (green plus). As megalin-dependent 25(OH)-vitaminD 3 endocytosis is impaired, low levels of the precursor vitamin D have access to its converting enzyme. Therefore, the delicate balance between the regulation of 1α-hydroxylase transcription (that cooperates to increase active vitamin D 3 levels) and the loss of the precursor form of vitamin D 3 into the urine (preventing its access to the enzyme), determine the outcome of either increased or decreased concentration of active vitamin D 3 in serum. Because vitamin D3 levels drives the absorption of calcium to the bloodstream, hypercalciuria and kidney stones may or may not develop.
higher endosomal Cl − concentration than a pure Cl − current . Unfortunately, the exact role of ClC-5 coupled Cl − /H + transport in early endosomes is still not fully understood.
## Clc-7: a lysosomal clc exchanger that requires a β-subunit
ClC-7 is another broadly expressed ClC protein. In mouse embryos, ClC-7 was found most prominently expressed in the brain, eyes, spinal cord, and dorsal root and trigeminal ganglia in mouse embryos [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] , whereas in adult mice it was found in medulla oblongata, Purkinje cells, PTCs, Sertoli cells, and pancreatic and tracheal epithelia [bib_ref] Localization of mouse CLC-6 and CLC-7 mRNA and their functional complementation of..., Kida [/bib_ref]. ClC-7 localizes mostly in lysosomes [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref] [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref] [bib_ref] Lysosomal degradation of endocytosed proteins depends on the chloride transport protein ClC-7, Wartosch [/bib_ref] , but is also found in the ruffled border of osteoclasts [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref]. ClC-7 is the only ClC exchanger that requires a β-subunit, Ostm1, for proper function. Ostm1 (osteopetrosisassociated membrane protein 1), a highly glycosylated type 1 transmembrane protein, is essential for stability and transport activity of ClC-7 [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref] [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref]. Mutations in the Ostm1 gene underlie the spontaneous gray-lethal mouse mutant [bib_ref] Grey-lethal mutation induces severe malignant autosomal recessive osteopetrosis in mouse and human, Chalhoub [/bib_ref]. Ostm1 and ClC-7 co-localize in lysosomes and in the ruffled border of osteoclasts and maintain a closely dependent relationship, in which protein levels of one are reduced by approximately 95% in the absence of the other [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref]. Moreover, Ostm1 needs to interact with ClC-7 in order to exit the ER and traffic to lysosomes, whereas ClC-7 needs Ostm1 to be stable and functional [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref] [bib_ref] Sorting motifs of the endosomal/lysosomal CLC chloride transporters, Stauber [/bib_ref]. The transmembrane domain of Ostm1 is necessary for ClC-7 trafficking to lysosomes, while the highly glycosylated N-terminus plays a critical role in transport activity of ClC-7 [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref].
For many years, the intracellular localization of CLC-7 has hindered the study of its biophysical properties. However, after the identification of a sorting motif localized at the cytosolic N-terminus that directs ClC-7 to lysosomes , point mutations that disrupt this motif allowed partial cell-surface localization of ClC-7 upon heterologous expression, allowing its biophysical characterization [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref]. ClC-7 shares several characteristics with other ClC exchangers such as the strong outward rectification; anion sequence conductance of Cl − > I − ; inhibition of activity upon low extracellular pH; and a classical 2Cl − /1H + stoichiometry. However, activation and deactivation of ClC-7 are very slow compared to other ClC transporters, allowing for the analysis of tail currents. Tail currents revealed that the exchange process is almost linearly voltage-dependent, and rectification is almost entirely due to a voltage gating [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref]. Later, slow voltage-dependent activation and deactivation of ClC-7 were assigned to the common gating mechanism [bib_ref] Common gating of both CLC transporter subunits underlies voltage-dependent activation of the..., Ludwig [/bib_ref]. ClC-7 also carries both gating and proton glutamates; mutation of these residues, such as is found in ClC-5, yields a protein displaying a Cl − conductance uncoupled from H + co-transport and a nonfunctional ClC-7 protein, respectively [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref].
## Clc-7 in osteopetrosis, retinal degeneration, and lysosomal storage disease
To study the physiological roles of ClC-7/Ostm1, knockout mouse models were generated and analyzed. ClC-7 KO mice present short life spans, severe osteopetrosis, retinal degeneration, lysosomal storage disease, and neurodegeneration [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref]. Gray-lethal mice (Ostm1 KO) display a very similar phenotype [bib_ref] Grey-lethal mutation induces severe malignant autosomal recessive osteopetrosis in mouse and human, Chalhoub [/bib_ref] [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref] , as expected for these two closely functionally related proteins. Interestingly, both ClC-7 KO and Ostm1 KO mice have gray fur in an agouti background (in which wild-type mice have brown fur), suggesting a possible role of ClC-7/Ostm1 in melanosomes [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref].
Loss of function of ClC-7 in osteoclasts results in osteopetrosis, a disease characterized by increased bone radiodensity because of ineffective osteoclast-mediated bone resorption [bib_ref] Osteopetrosis: current clinical considerations, Shapiro [/bib_ref]. The ruffled border of osteoclasts-a membrane domain responsible for acidic digestion of bone tissue-is formed by lysosomal membrane insertion and exocytosis of their content. Acidification of the resorption lacuna-the space between the ruffled border and the bone tissue-is carried by V-type H + -ATPase that, similarly to compartments of the endosomal/lysosomal pathway, requires an electrical shunt thought to be performed by ClC-7/Ostm1 (Planells-Cases and [bib_ref] Cell biology and physiology of CLC chloride channels and transporters, Stauber [/bib_ref]. In the resorption lacuna, ClC-7/Ostm1 is responsible for the Cl − influx necessary for neutralization (shunting) of protons, which keeps the proton pump functional and maintains the lacuna's highly acidic environment . Indeed, osteoclasts from ClC-7 KO mice displayed underdeveloped ruffled borders and impaired acidification of the resorption lacuna, which causes the osteopetrotic phenotype [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref].
After establishment of the link between ClC-7/Ostm1 mutations and osteopetrosis, approximately 50 different human ClC-7 mutations were identified in osteopetrotic patients (for details, see [bib_ref] Cell biology and physiology of CLC chloride channels and transporters, Stauber [/bib_ref]. Most of these are missense mutations, some of which cause an autosomal dominant form of osteopetrosis that presents less severe symptoms and does not implicate the nervous system [bib_ref] Albers-schonberg disease (autosomal dominant osteopetrosis, type II) results from mutations in the..., Cleiren [/bib_ref] [bib_ref] Chloride channel ClCN7 mutations are responsible for severe recessive, dominant, and intermediate..., Frattini [/bib_ref]. Others yield ClC-7/Ostm1 complexes, which are retained in the ER or strongly reduce their ion transport activity [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref]. However, other missense mutations produce transporters that reach the normal subcellular localization, remaining functional when expressed heterologously in the plasma membrane, and carrying the sorting motif point mutation mentioned above. Curiously, these functional mutants yielded currents with accelerated kinetics of activation and deactivation. Given that patients carrying these mutations present symptoms similar to others with non-functional transporters, the slow gating of ClC-7/Ostm1 is likely physiologically important [bib_ref] ClC-7 is a slowly voltage-gated 2Cl(-)/1H(+)-exchanger and requires Ostm1 for transport activity, Leisle [/bib_ref].
In addition to osteopetrosis, ClC-7 KO and gray-lethal mice also display retinal and neurodegeneration associated with lysosomal storage [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref] [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref]. Retinal degeneration leads to blindness about 4 weeks after birth [bib_ref] Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and..., Kornak [/bib_ref] [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref]. Although previously the blindness was believed to be the result of osteopetrotic narrowing of the optic nerve canal [bib_ref] Neurological aspects of osteopetrosis, Steward [/bib_ref] , retinal degeneration is a direct effect of disruption of ClC-7 or Ostm1 at retinal neurons [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref]. Neurodegeneration is the probable cause of death of ClC-7 FIGURE 7 | ClC-7 function at osteoclasts. Lysosomes are inserted into the 'ruffled border' of bone-attached osteoclasts. The resorption lacuna is then acidified by the combined work of proton pumps (pink) and ClC-7/Ostm1 exchangers (green) present in the lysosomes membranes. Low pH conditions are required for the chemical dissolution of inorganic bone material and for the activity of lysosomal enzymes that are secreted into the lacuna.
KO mice at approximately 6 weeks. They present neuronal cell loss, particularly in the hippocampus and cerebral cortex, as well as lysosomal storage material scattered throughout the neuronal somata [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref]. Neurodegeneration and lysosomal storage disease are cell-autonomous effects of disruption of ClC-7 as demonstrated by tissue-specific ClC-7 KO mice. In this study, PTCs and neurons lacking or expressing ClC-7 were compared within the same environment; only cells devoid of ClC-7 displayed lysosomal disease [bib_ref] Lysosomal degradation of endocytosed proteins depends on the chloride transport protein ClC-7, Wartosch [/bib_ref].
## New insights on the role of c1c-7 in lysosomes
In correlation with ClC-7 function in osteoclasts and ClC-3 to ClC-5 roles in acidification of their respective compartments [bib_ref] The chloride channel ClC-4 contributes to endosomal acidification and trafficking, Mohammad-Panah [/bib_ref] , the lysosomal storage phenotype was first proposed to result from impaired acidification. However, ratiometric measurements showed that ClC-7/Ostm1 does not play a role in lysosomal acidification; that is, lysosomes from ClC-7 KO mice display normal steady-state pH [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref] [bib_ref] ClC-7 requires Ostm1 as a β-subunit to support bone resorption and lysosomal..., Lange [/bib_ref] [bib_ref] Lysosomal pathology and osteopetrosis upon loss of H+-driven lysosomal Cl-accumulation, Weinert [/bib_ref]. The counter-charge conductance suggested to neutralize H + -ATPase currents in this case seems to be provided by a lysosomal cation efflux [bib_ref] A cation counterflux supports lysosomal acidification, Steinberg [/bib_ref].
These intriguing results concerning the function of ClC-7 in lysosomes led to the generation of two other ClC-7 mice models, the ClC-7 unc mice -generated similarly to the ClC-5 unc mice, by a 'gating glutamate' mutation that yields uncoupled Cl − pure conductance-and the ClC-7 td mice [bib_ref] Transport activity and presence of ClC-7/Ostm1 complex account for different cellular functions, Weinert [/bib_ref] , generated by a mutation in the 'proton glutamate' that abolishes transport (td: transport deficient). Much like ClC-7 KO mice, both new models displayed normal lysosomal pH and reduced Cl − concentration. However, although ClC-7 unc mice presented neuronal cell loss and lysosomal storage disease similar to the ClC-KO mice, the osteopetrotic phenotype was partially rescued by the presence of pure Cl − currents . Meanwhile, ClC-7 td mice displayed similar osteopetrotic phenotype to ClC-7 KO mice, but a delayed and less severe neurodegeneration [bib_ref] Transport activity and presence of ClC-7/Ostm1 complex account for different cellular functions, Weinert [/bib_ref].
Surprisingly, unlike ClC-7 KO and Ostm1 KO mice, the two new models presented normal agouti brown fur color [bib_ref] Transport activity and presence of ClC-7/Ostm1 complex account for different cellular functions, Weinert [/bib_ref]. Taken together, these results suggest that the physical presence of the non-functioning ClC-7 td alone was sufficient to rescue the normal fur pigmentation and alleviate neurodegeneration, whereas ClC-7 unc Cl − currents had a positive effect in partially rescued osteopetrosis and normal fur pigmentation, but a negative effect in neurodegeneration [bib_ref] Transport activity and presence of ClC-7/Ostm1 complex account for different cellular functions, Weinert [/bib_ref]. Therefore, normal fur pigmentation seems to require only the presence of ClC-7/Ostm1 complex, whereas lysosomes and osteoclasts require fully functional Cl − /H + exchangers in order to function properly. The precise role of ClC-7 in lysosomes is still obscure, as is the previously discussed role of ClC-5 in endosomes. More work is necessary to better understand the role of coupled Cl − /H + transport in the endosomal/lysosomal pathway.
## Clcs with uncertain physiological function
The last sections of this review focuses on ClC-3, -4, and -6, which have uncertain physiological functions. Although mutations or dysfunction of these ClCs have not been linked directly to specific human diseases, mouse models have demonstrated important phenotypes that are worth discussing here.
## Clc-3: retinal and brain degeneration?
ClC-3 is a broadly expressed intracellular ClC protein with controversial biophysical and physiological characteristics. Several mutually conflicting Cl − currents have been attributed to ClC-3. First, a slightly outwardly rectifying Cl − current inhibited by PKC [bib_ref] Cloning and expression of a protein kinase C-regulated chloride channel abundantly expressed..., Kawasaki [/bib_ref] ; second, Cl − currents inhibited by intracellular Ca +2 [bib_ref] Stable and functional expression of the CIC-3 chloride channel in somatic cell..., Kawasaki [/bib_ref] , followed by swelling-activated Cl − currents, also known as VRACs [bib_ref] Molecular identification of a volume-regulated chloride channel, Duan [/bib_ref]. The role of VRACs was not confirmed by data from three independently generated ClC-3 KO mice [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref] [bib_ref] Altered GABAergic function accompanies hippocampal degeneration in mice lacking ClC-3 voltage-gated chloride..., Dickerson [/bib_ref] [bib_ref] CLC-3 deficiency leads to phenotypes similar to human neuronal ceroid lipofuscinosis, Yoshikawa [/bib_ref] , as swelling-activated Cl − currents were unaffected in the hepatocytes, salivary acinar cells and cardiomyocytes [bib_ref] Secretion and cell volume regulation by salivary acinar cells from mice lacking..., Arreola [/bib_ref] [bib_ref] ClC-3-independent, PKC-dependent activity of volumesensitive cl channel in mouse ventricular cardiomyocytes, Gong [/bib_ref]. Moreover, a recent report identified LRRC8 proteins as essential components of VRACs, as their disruption abolishes VRAC currents [bib_ref] Identification of LRRC8 heteromers as an essential component of the volumeregulated anion..., Voss [/bib_ref].
ClC-3 presents very low cell surface expression, even when heterologously overexpressed, which has hindered a thorough analysis of its biophysical properties. Nevertheless, in some studies, small but strongly outwardly rectifying Cl − currents were found [bib_ref] The ClC-3 chloride channel promotes acidification of lysosomes in CHO-K1 and huh-7..., Li [/bib_ref] [bib_ref] Overexpression of CLC-3 in HEK293T cells yields novel currents that are pH..., Matsuda [/bib_ref] [bib_ref] ClC-3 is an intracellular chloride/proton exchanger with large voltage-dependent nonlinear capacitance, Guzman [/bib_ref] , similar to those reported for ClC-4 and ClC-5 [bib_ref] Cloning and functional expression of rat CLC-5, a chloride channel related to..., Steinmeyer [/bib_ref] [bib_ref] Mutational analysis demonstrates that ClC-4 and ClC-5 directly mediate plasma membrane currents, Friedrich [/bib_ref]. Moreover, because of its substantial sequence identity-approximately 80% with ClC-4 and ClC-5, which are well established ClC exchangers-and the presence of the conserved 'gating glutamate, ' which mutation abolish rectification as in other ClCs, ClC-3 is most likely a voltage dependent Cl − /H + exchanger [bib_ref] The ClC-3 chloride channel promotes acidification of lysosomes in CHO-K1 and huh-7..., Li [/bib_ref] [bib_ref] ClC-3 is an intracellular chloride/proton exchanger with large voltage-dependent nonlinear capacitance, Guzman [/bib_ref].
ClC-3 is expressed in most tissues, including the brain, retina, adrenal gland, pancreas, intestines, epididymis, kidney, liver, skeletal muscle, and heart [bib_ref] Cloning and expression of a protein kinase C-regulated chloride channel abundantly expressed..., Kawasaki [/bib_ref] [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref] [bib_ref] Intracellular localization of ClC chloride channels and their ability to form hetero-oligomers, Suzuki [/bib_ref] [bib_ref] Role of the vesicular chloride transporter ClC-3 in neuroendocrine tissue, Maritzen [/bib_ref]. It mainly resides in endosomes, where it was found co-localized with ClC-4 and ClC-5, and also with both early and late endosomal markers [bib_ref] Intracellular localization of ClC chloride channels and their ability to form hetero-oligomers, Suzuki [/bib_ref]. CLC-3 is also found in synaptic vesicles [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref] [bib_ref] AP-3-dependent mechanisms control the targeting of a chloride channel (ClC-3) in neuronal..., Salazar [/bib_ref] and synaptic-like micro vesicles [bib_ref] AP-3-dependent mechanisms control the targeting of a chloride channel (ClC-3) in neuronal..., Salazar [/bib_ref] [bib_ref] Role of the vesicular chloride transporter ClC-3 in neuroendocrine tissue, Maritzen [/bib_ref]. Three different ClC-3 KO mouse lines displayed similar phenotypes of severe degeneration of the retina and brain, with prominent effects in the hippocampus [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref] [bib_ref] Altered GABAergic function accompanies hippocampal degeneration in mice lacking ClC-3 voltage-gated chloride..., Dickerson [/bib_ref] [bib_ref] CLC-3 deficiency leads to phenotypes similar to human neuronal ceroid lipofuscinosis, Yoshikawa [/bib_ref]. In one model, signs of lysosomal storage disease were observed [bib_ref] CLC-3 deficiency leads to phenotypes similar to human neuronal ceroid lipofuscinosis, Yoshikawa [/bib_ref] , but these effects were much weaker than those found in ClC-6 and ClC-7 KO mice [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref] [bib_ref] Loss of the chloride channel ClC-7 leads to lysosomal storage disease and..., Kasper [/bib_ref]. The mechanism by which ClC-3 causes neurodegeneration is still unclear.
Although ClC-3 was thought to provide the electrical shunt for acidification of intracellular compartments like the other ClC exchangers, its role in endosomes and synaptic vesicles is still controversial. In ClC-3 KO mice, acidification and Cl − accumulation were reduced in early and late endosomes [bib_ref] ClC-3 chloride channels facilitate endosomal acidification and chloride accumulation, Hara-Chikuma [/bib_ref] and synaptic vesicles showed less efficient acidification in vitro [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref]. Synaptic vesicles from ClC-3 KO mice exhibit reduced glutamate uptake, but this feature has been ascribed to diminished levels of the vesicular glutamate transporter VGLUT1 [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref]. In another study, evidence against ClC-3's role in acidifying synaptic vesicles was reported [bib_ref] A chloride conductance in VGLUT1 underlies maximal glutamate loading into synaptic vesicles, Schenck [/bib_ref]. The fact that VGLUT1 KO mice display no chloride-dependent acidification of synaptic vesicles, and very little expression of CLC-3 in synapse vesicles, led the authors to suggest that VGLUT1 represents the major Cl − conductance pathway in synaptic vesicles [bib_ref] A chloride conductance in VGLUT1 underlies maximal glutamate loading into synaptic vesicles, Schenck [/bib_ref]. Therefore, the reduced acidification of synaptic vesicle reported in ClC-3 KO mice [bib_ref] Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to..., Stobrawa [/bib_ref] was attributed to reduced levels of VGLUT1, which itself most likely resulted from severe neurodegeneration.
To avoid the effects of neurodegeneration on VGLUT1 levels, [bib_ref] Involvement of ClC-3 chloride/proton exchangers in controlling glutamatergic synaptic strength in cultured..., Guzman [/bib_ref] used cultured hippocampal neurons to analyze ClC-3 disruption in synaptic vesicles. The authors observed enlargement of synaptic vesicles and increased glutamate content in cells lacking ClC-3. The probability of vesicle fusion and release of its content was also increased in those cells, indicating that exaggerated release of glutamate in the synaptic cleft contributes to neurodegeneration in ClC-3 KO mice [bib_ref] Involvement of ClC-3 chloride/proton exchangers in controlling glutamatergic synaptic strength in cultured..., Guzman [/bib_ref].
Recently, reduced levels of ClC-3 were found in patients with inflammatory bowel disease (IBD) and from mice treated with dextran sulfate sodium (DSS) to induce colitis and mimic IBD. ClC-3 KO mice were more susceptible to DSS-induced colitis with no signs of recovery after treatment. Lack of ClC-3 provoked apoptosis of intestinal epithelial cells, causing disruption of the epithelial barrier and bacterial invasion. Thus, the authors defend the involvement of ClC-3 in IBDs pathogenesis. Another study points to ClC-3's involvement atherosclerosis. In an ApoE null mice background, further disruption of ClC-3 reduced the size of atherosclerotic lesions present in the aorta. The authors suggest that ClC-3 insufficiency disrupts scavenger receptor SR-A expression (via JNK/p38 MAPK) and foam cell formation, leading to reduction/inhibition of atherosclerotic lesions [bib_ref] ClC-3 deficiency prevents atherosclerotic lesion development in ApoE-/-mice, Tao [/bib_ref]. The protective effect of ClC-3 deficiency was again addressed in endothelial progenitor cells (EPCs). Angiotensin II-induced apoptosis of EPCs was remarkably reduced in ClC-3 KO mice. This inhibition was attributed to suppressed levels of reactive oxygen species and NADPH oxidase activity-direct effects of angiotensin-II-that are suppressed by ClC-3 disruption [bib_ref] ClC-3 deficiency prevents apoptosis induced by angiotensin II in endothelial progenitor cells..., Liu [/bib_ref].
Despite ClC-3 diversity in different cell types, phenotypes present in ClC-3 KO mice strongly suggest that ClC-3 is important for neurotransmission in the CNS. Three different splicing variants of ClC-3 (ClC-3a, ClC-3b, and ClC-3c) were described in the brain with different subcellular localization but similar transport function [bib_ref] Neuronal ClC-3 splice variants differ in subcellular localizations, but mediate identical transport..., Guzman [/bib_ref]. Another splicing variant, ClC-3d, was described in mouse livers as displaying different localizations but identical transport properties [bib_ref] A newly cloned ClC-3 isoform, ClC-3d, as well as ClC-3a mediates cd-sensitive..., Okada [/bib_ref]. The number of splicing variants with different subcellular localizations might explain the diversity of functions ascribed to ClC-3; the study of these isoforms could be a promising direction for further study of the precise function and localization of ClC-3 proteins, which could provide an explanation for the phenotypes described in KO mice.
## Clc-4: a cl − /h + exchanger with unclear physiological function
ClC-4 is a broadly expressed ClC exchanger found in various tissues which differ between species. It is found mostly in the muscles, brain, and heart of humans [bib_ref] A gene from the Xp22.3 region shares homology with voltage-gated chloride channels, Van Slegtenhorst [/bib_ref] ; in the liver and brain, heart, muscles, spleen, and kidneys of rats ; and in the brain, intestines, and kidneys of mice [bib_ref] The chloride channel ClC-4 contributes to endosomal acidification and trafficking, Mohammad-Panah [/bib_ref]. Interestingly, the gene encoding ClC-4 is localized on chromosome 7 in inbred laboratory mice, but in humans and rats, the gene resides on the X chromosome [bib_ref] Different chromosomal localization of the Clcn4 gene in mus spretus and C57BL/6J..., Rugarli [/bib_ref]. This may partially explain the variety and species-specificity of expression patterns.
ClC-4 localizes mainly at endosomes' membranes; upon heterologous overexpression, a small portion is also found within the plasma membrane [bib_ref] The chloride channel ClC-4 contributes to endosomal acidification and trafficking, Mohammad-Panah [/bib_ref] [bib_ref] Intracellular localization of ClC chloride channels and their ability to form hetero-oligomers, Suzuki [/bib_ref]. Like other members of this family, the cell surface localization allows for a better analysis of its biophysical properties. ClC-4 yields a strongly outwardly rectifying current that is inhibited by low extracellular pH; to date, the physiological relevance of pH regulation remains unclear. ClC-4 is a Cl − /H + exchanger with anion conductance sequence of Cl − > Br − > I − , similar to other intracellular ClCs. Mutation of the 'gating glutamate' strongly changes rectification and converts the exchanger into a passive Cl − conductor, highlighting the importance of this residue in coupling proton to Cl − transport [bib_ref] Mutational analysis demonstrates that ClC-4 and ClC-5 directly mediate plasma membrane currents, Friedrich [/bib_ref] [bib_ref] Chloride/ proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5, Picollo [/bib_ref] [bib_ref] Voltage-dependent electrogenic chloride/ proton exchange by endosomal CLC proteins, Scheel [/bib_ref]. [bib_ref] Channel-like slippage modes in the human anion/proton exchanger ClC-4, Alekov [/bib_ref] have shown that ClC-4 proteins, when exposed to different types of anions in the extracellular buffer, can display a phenomenon called 'slippage, ' where the transporter behaves as a channel rather than an obligatory exchanger. High extracellular SCN − cause increased current amplitudes and uncouple H + transport rendering ClC-4 channellike transport with reduced H + currents and biophysical properties similar to other well-characterized ClC channels. Restoring extracellular Cl − rescues Cl − /H + exchange. In ClC-5, extracellular SCN − uncouples transport but does not affect proton transport [bib_ref] Glutamate 268 regulates transport probability of the anion/proton exchanger ClC-5, Grieschat [/bib_ref] , an apparent isoform-specific effect. Thus, ClC-4 is suggested to function as a channel or exchanger depending on the extracellular anion [bib_ref] Channel-like slippage modes in the human anion/proton exchanger ClC-4, Alekov [/bib_ref].
Currently, there is little scientific consensus regarding the subcellular localization of ClC-4 . On those studies, ClC-4 was found in sub-apical vesicles of proximal tubule epithelium [bib_ref] The chloride channel ClC-4 contributes to endosomal acidification and trafficking, Mohammad-Panah [/bib_ref] ; in intracellular compartments of HEK 293 cells co-localizing with ClC-3 and ClC-5 [bib_ref] Intracellular localization of ClC chloride channels and their ability to form hetero-oligomers, Suzuki [/bib_ref] ; and in the endoplasmic reticulum [bib_ref] Human ClC-4 protein, a member of the CLC chloride channel/transporter family, is..., Okkenhaug [/bib_ref]. However, none of the immunohistochemistry studies performed thus far have used cells from ClC-4 KO mice as a negative control to confirm their data.
ClC-4 was suggested to facilitate endosomal acidification by working as the electrical shunt for proton accumulation mediated by the proton pump. However, ClC-4 KO mice do not display any obvious abnormal phenotypes . Although ClC-4 trafficking is similar to ClC-5, they do not appear to perform similar physiological functions [bib_ref] The chloride channel ClC-4 contributes to endosomal acidification and trafficking, Mohammad-Panah [/bib_ref]. The additional disruption of ClC-4 in ClC-5 KO mice did not aggravate the impaired endocytosis phenotype in PTCs .
One naturally occurring mutation (G544R), found in a patient with severe epilepsy and delayed development, nearly abolished ClC-4 currents when expressed heterologously [bib_ref] Exome sequencing reveals new causal mutations in children with epileptic encephalopathies, Veeramah [/bib_ref]. [bib_ref] X-exome sequencing of 405 unresolved families identifies seven novel intellectual disability genes, Hu [/bib_ref] , analyzing X-linked intellectual disabilities, identified five different mutations in the ClC-4 gene in five families. Currents of ClC-4 proteins carrying each of these mutations were much smaller or even absent compared to wildtype ClC-4. Moreover, C1C-4 depletion in cultured hippocampal neurons, affected neuronal differentiation; the cells displayed a 30% reduction of neuritic outgrowth and branching . Additional studies using specific antibodies and appropriate KO controls are necessary to further understand ClC-4 physiological function in specific cell compartments, determine precise sub-cellular localization, and investigate possible roles in human diseases.
## Clc-6: mild lysosomal storage disease?
ClC-6 shares approximately 45% of its sequence identity with ClC-7; together, they form the third branch of the ClC protein family. Like other ClC exchangers, ClC-6 localizes at membranes of the endosomal/lysosomal pathway [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref]. ClC-6 mRNA was found in several tissues [bib_ref] ClC-6 and ClC-7 are two novel broadly expressed members of the CLC..., Brandt [/bib_ref] , but the expressed ClC-6 protein is found almost exclusively in the nervous system [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref]. First attempts to record ClC-6 currents by heterologous expression were frustrated by its late endosomal localization [bib_ref] ClC-6 and ClC-7 are two novel broadly expressed members of the CLC..., Brandt [/bib_ref] [bib_ref] Evidence for the intracellular location of chloride channel (ClC)-type proteins: Co-localization of..., Buyse [/bib_ref] , and biophysical characterization only became possible when GFP-tagged ClC-6 proteins were expressed in the plasma membrane [bib_ref] The late endosomal ClC-6 mediates proton/chloride countertransport in heterologous plasma membrane expression, Neagoe [/bib_ref]. ClC-6 mediates outwardly rectifying currents that are reduced by extracellular acidification, as in other ClC-exchangers. Mutation in the 'gating glutamate' also disrupts rectification, turning ClC-6 into a passive Cl − conduit [bib_ref] The late endosomal ClC-6 mediates proton/chloride countertransport in heterologous plasma membrane expression, Neagoe [/bib_ref].
Knock-out controlled immunohistochemistry studies have shown that native ClC-6 localizes predominantly at late endosomes of neurons in situ [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref] and in cultured cells [bib_ref] Human ClC-6 is a late endosomal glycoprotein that associates with detergent-resistant lipid..., Ignoul [/bib_ref] , whereas in heterologous expression it is also found co-localized with early and late endosomal markers [bib_ref] Intracellular localization of ClC chloride channels and their ability to form hetero-oligomers, Suzuki [/bib_ref] [bib_ref] Sorting motifs of the endosomal/lysosomal CLC chloride transporters, Stauber [/bib_ref].
ClC-6 KO mice present no apparent abnormal phenotypes, with normal life span and weight. However, late in life (>3 months old), the mice display a peculiar form of lysosomal storage disease, with deposits found in central and peripheral neurons [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref]. Different from ClC-7 KO mice, in which such deposits are localized all over the neuronal soma and the disease progression is much more aggressive, deposits in ClC-6 KO neurons are mainly localized at initial axon segments and the disease progresses very slowly [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref] [bib_ref] Distinct neuropathologic phenotypes after disrupting the chloride transport proteins ClC-6 or ClC-7/Ostm1, Pressey [/bib_ref]. Moreover, the absence of ClC-6 in hippocampal neurons does not affect lysosomal steady-state pH [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref]. Deposits found in ClC-6 KO mice tested positive for markers typically found in neuronal ceroid lipofuscinosis (NCL), a lysosomal storage disease. The authors therefore proposed ClC-6 gene as a candidate for mild forms of NCL, but did not find convincing association upon analysis of 75 NCL patients [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref].
In general, neuropathology in ClC-6 KO mice is much milder than in ClC-3 and ClC-7 KO mice. They show no vision impairment, and little neuronal cell loss and microglial activation [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref] [bib_ref] Distinct neuropathologic phenotypes after disrupting the chloride transport proteins ClC-6 or ClC-7/Ostm1, Pressey [/bib_ref]. ClC-6 KO mice also demonstrate reduced pain sensitivity, correlated with an impairment of dorsal root ganglion neuronal function due to dramatic lysosomal storage accumulation [bib_ref] Lysosomal storage disease upon disruption of the neuronal chloride transport protein CIC-6, Poët [/bib_ref]. After all, like ClC-3 and ClC-4, ClC-6 is another ClC exchanger whose physiological role is poorly understood at present.
# Conclusion
Cl − ion transport has risen from obscurity to become a vibrant and exciting field in ion transport research. Within this field, ClC proteins are a particularly intriguing family of anion channels and transporters involved in several important physiological functions. Twenty-five years after the discovery of its first member (ClC-0), and following enormous efforts to study their biological aspects, many questions about the structure, function, and pathophysiological roles of ClCs have been answered, but an equally high number of new and, so far, unsolved questions have emerged. For instance, the precise localization of ClC-K channels in the thin limb of the loop of Henle in the kidney and its function in intercalated cells are still unknown. Future research topics of particular interest include a better understanding of the relationship between αand β-subunits, and of the physiological role of β-subunits by themselves.
Phenotypes of mouse models have linked ClC protein function and dysfunction with inherited human genetic diseases. Myotonia congenita, leukodystrophy, Bartter syndrome, Dent's disease, and osteopetrosis/retinal degeneration/lysosomal storage disease have well-established association with loss-of-function of ClC1, ClC-2, ClC-K/Barttin, ClC-5 and ClC-7/Ostm1, respectively. However, many aspects of these diseases' molecular origins remain obscure.
Useful tools to increase our knowledge about the molecular basis of ClC-related diseases would include the development of small molecules able to specifically block or activate ClC proteins. Unfortunately, currently available compounds targeting ClC proteins are few and far between, and they lack specificity. The role of intracellular ClC exchangers in the endosomal/lysosomal pathway is not completely established. Acidification and Cl − accumulation seem not to be the only functions of ClC exchangers in these compartments. Interactions with other cell proteins-and not only transport activity of ClC exchangersseem to play a role in endosome/lysosome homeostasis, as revealed by the rescue of some pathological phenotypes in KO mice upon the expression of non-functional ClC-7 proteins. Some phenotypes displayed by ClC-3 and ClC-6 KO mice could not be correlated with the physiological roles so far assigned to ClC-3 and ClC-6. However, this may be only a matter of time; recently, leukodystrophy and azoospermia-typically phenotypes of ClC-2 KO mice-were described in patients with ClC-2 mutations.
Crystal structures of prokaryote and eukaryote ClC proteins have provided important insights about molecular structure and ion conductance mechanisms. ClC proteins are unique in their double-barreled structure, providing a new model of ion transport in which the same basic architecture supports bona fide channel conductance and ion co-transport. These two types of ion translocation were believed to occur by entirely distinct mechanisms. However, the available crystal structures were not sufficient to uncover the molecular mechanism governing the common gating mechanism and the precise proton transport pathway. Use of new approaches or the development of novel techniques may be necessary to uncover the molecular mechanisms underlying ClC ion transport.
Generation of crystal structures of mammalian ClC channels and exchangers will ultimately permit a more accurate investigation into the differences between these two structures, and also the identification of regions involved in interaction and modulation by other cellular components. Moreover, those structures will greatly assist in the development of new compounds able to modify specific types of ClC proteins, thus opening the field for pharmacological approaches aiming at generating therapeutic drugs. Such drugs would have the potential to reduce or even eliminate the undesired symptoms caused by ClC proteins loss-of-function, improving quality of life for many patients.
# Author contributions
DP analyzed the literature, wrote the paper, and prepared the figures; RP analyzed the literature and reviewed the paper; VC analyzed the literature, reviewed the paper and supervised the work.
FUNDING DP is a Science without border-Brazil and Nova Scotia Graduate Scholar. VC is a Cystic Fibrosis Canada researcher.
[table] TABLE 1 |: Mammalian ClC chloride channels. [/table]
[table] TABLE 2 |: Mammalian ClC chloride exchangers. [/table]
|
Nanoscale gizmos – the novel fluorescent probes for monitoring protein activity
a b s t r a c t Nanobiotechnology has emerged inherently as an interdisciplinary field, with collaborations from researchers belonging to diverse backgrounds like molecular biology, materials science and organic chemistry. Till the current times, researchers have been able to design numerous types of nanoscale fluorescent tool kits for monitoring protein-protein interactions through real time cellular imagery in a fluorescence microscope. It is apparent that supplementing any protein of interest with a fluorescence habit traces its function and regulation within a cell. Our review therefore highlights the application of several fluorescent probes such as molecular organic dyes, quantum dots (QD) and fluorescent proteins (FPs) to determine activity state, expression and localization of proteins in live and fixed cells. The focus is on Fluorescence Resonance Energy Transfer (FRET) based nanosensors that have been developed by researchers to visualize and monitor protein dynamics and quantify metabolites of diverse nature. FRET based toolkits permit the resolution of ambiguities that arise due to the rotation of sensor molecules and flexibility of the probe. Achievements of live cell imaging and efficient spatiotemporal resolution however have been possible only with the advent of fluorescence microscopic technology, equipped with precisely sensitive automated softwares.
# Introduction
Fluorescence is a luminescence process in which atoms and molecules absorb a specific lower wavelength light and after a brief gap of fluorescence lifetime emit a longer wavelength light. Over the past many years in the service of biology, fluorescence microscopy has emerged as the prime pillar of microscopy due to its inherent selectivity to expose only objects of interest against black background [bib_ref] Fluorescence microscopy, Lichtman [/bib_ref]. To investigate myriad of cellular processes, fluorescence imaging has been realistic for the visualization of molecules and whole organisms. It started with the attachment of organic dyes to proteins through antibodies and later sought the genetic tagging of target proteins with fluorescent proteins. However, the linking of antibodies has to be supplemented with the fixation and permeabilization of cells [bib_ref] The fluorescent toolbox for assessing protein location and function, Giepmans [/bib_ref]. With the efficacy of fluorophores as direct recognition agents for various cellular molecules like nucleic acids, ions and other cell organelles, the technique of immuno-fluorescence became less valuable. Additionally, with the development of FRET based sensors, noninvasive behavior and live cell imaging in both prokaryotic and eukaryotic cells proved to be a novelty for researchers exploring fluorescent probes for real time cellular imaging. Originally, FRET stands for Forster Resonance Energy Transfer in the honour of Theodor Forster, physical chemist, who first discovered and understood it. The term "fluorescence resonance energy transfer" is often used, when both the chromophores are fluorescent. The latter enjoys common usage in scientific literature and has been in practice as such commonly.
However, It has been established that compared to other fluorophores, amplified brightness and photo-stability are the two critical parameters of semiconductor nanocrystals that makes them unique for fluorescent microscopic imaging [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref] , although their targeting potential still endures many hindrances. In recent years, using newly engineered photo-controllable fluorescent proteins; numerous super-resolution microscopic techniques became ready to use in order to get better visualization of objects having dimensions smaller than 500 nm and 200 nm in axial and lateral directions, respectively. In current times, researchers are now powerfully aided with more precisely optimized FPs. There are three groups of photo-controllable FPs-photo-convertible FPs (PCFPs), photoactivatable FPs (PAFPs), and reversibly photoswitchable FPs (rsFPs). PAFPs are activated from non-fluorescent (dark) to fluorescent state, whereas, rsFPs are reversibly photoswitched between inactive and active states, and PCFPs are made to convert from original color to another. The super-resolution microscopic imaging using these FPs is always carried out by controllably turning them on and off [bib_ref] Photocontrollable fluorescent proteins for superresolution imaging, Shcherbakova [/bib_ref].
Moreover, proteases have emerged favorable enzymes for synthesizing their inhibitory compounds as potential drugs for some major human diseases such as cancer, AIDS and other neurodegenerative infections. Because of the latest coherent advancements between genomic, proteomic and biophysical techniques, peptide substrates for several proteases have been ascertained, that paved a way to trace and analyze the functioning of their equivalent proteases through a simplistic and speedy mode. By combining such peptide substrates with appropriate FPs, researchers develop a chimeric protein and then sequentially analyze many protease inhibitors through FRET disruption. Present review will focus on latest advances in major fluorophores and diverse fluorescence strategies that are in current use of fluorescence microscopy to visualize protein dynamics involved in their location and function within a cell.
## Variety of fluorescent probes
Varieties of fluorophores that are beyond the scope of this review are now available for evaluating the protein activity.
Regarding fluorescent probes, two critical considerations must be acknowledged first the fact that spectral properties of fluorescent probes determine the basic settings of time resolution and wavelength of instruments, and secondly certain fluorescent probes are used to monitor specific activities. For example, fluorophores that are sensitive to pH can only be explored to measure pH, and rotational diffusion can be tracked by only those probes having non-zero anisotropies. For histological studies, probes possessing long excitation and emission wavelengths are utilized to display auto-fluorescence at short excitation wavelengths. We are elaborating here some basic fluorophores which are being exploited through various bioengineering approaches to visualize the sea of protein dynamics, so as to come up with innovative nanoscale toolkits for monitoring molecular interactions operating within a living cell in real time.
## Molecular organic dyes
In the synthesis of organic fluorophores, many molecular strategies have been adapted. For this purpose, addition of electrophiles or other charged substituents like sulfonates, conjugation of double bonds, extra ring rigidification by locking rotatable rings into parent ring structures, etc. are some of the most important assignments. Though the dyes made by such strategies are now available commercially , however, such dyes are handicapped by being non-specific to all proteins. Therefore, their use in permeabilized and fixed cells has to be supplemented with antibodies [fig_ref] Figure 2: Application of different types of fluorophores, exploited for labeling and detection of... [/fig_ref]. These organic fluorophores are usually of low magnitude (<1 kD) and their important etiquettes like reduction in self-quenching, good brightness and appropriate wavelength have made them industrially optimized.
For a successful delivery of fluorophore into a cell, broadly two types of cellular environments-intracellular and extracellular are available. The major confrontations to overcome lies in optimized intracellular delivery of fluorophores, the perfect labeling of target biomolecule and negligible cytotoxicity. Due to smaller size of organic fluorophores there is least spatial hindrance to impede with functioning of target biomolecule. Therefore, this opportunity has been explored by attaching many fluorescent probes with a single target biomolecule in order to gather maximum fluorescence signal. Because of high label densities the strong electrostatic repulsion between neighboring molecules, the dye structural conformation and hydrophilicity altogether can cause fluorescence quenching [bib_ref] Cyanine dye labeling reagents: sulfoindocyanine succinimidyl esters, Mujumdar [/bib_ref] [bib_ref] Anomalous fluorescence enhancement of Cy3 and Cy3. 5 versus anomalous fluorescence loss..., Gruber [/bib_ref] and may also affect functioning of the biomolecule [bib_ref] Quantum dots versus organic dyes as fluorescent labels, Resch-Genger [/bib_ref].
For the selective detection of enzymatic and non-enzymatic proteins, fluorescent chemical turn-on probes are synthesized by fusing an environment sensitive fluorophore with a protein-specific small molecule [bib_ref] Environment-sensitive fluorescent turn-on probes targeting hydrophobic ligand-binding domains for selective protein detection, Zhuang [/bib_ref]. In this scenario, localized deposition of proteins in proteostasis has been elaborately exploited to develop sensors for understanding the mechanism of protein deposition in neurodegenerative disease progression [bib_ref] Adapting proteostasis for disease intervention, Balch [/bib_ref]. The process of proteostatis involves biogenesis, traffic and protein degradation within and outside a cell, involving different integrated biological pathways. The basis of such a designed fluorescent probe is that most of the ligand binding sites in proteins are hydrophobic, that constitute the prime thermodynamic driving force for the binding of small molecule ligands to their respective proteins. This approach has been exploited to develop a precious turn on folding sensor for effective live cell monitoring of proteostatis [bib_ref] Fluorescence turn-on folding sensor to monitor proteome stress in live cells, Liu [/bib_ref]. This small fluorogenic molecule became fluorescent when it binds and reacts with folded and functional retroaldolase enzyme. Recently, in a similar approach, dual signal fluorescence-enhanced sensor, based on Cu 2+ mediated fluorescence switchable strategy has been designed to detect Cysteine (Cys) in a simple and fast way. It was observed that two fluorescence emissions of ultrathin films (calcein@NFR/LDHs UTFs) are effec- tively quenched by Cu 2+ (off state), and then reversibly recovered by Cys (on state), leading to the specific coordination of Cys and Cu 2+. Similarly, a novel, simple and rapid fluorescent probe based on excited-state intramolecular proton transfer (ESIPT) was the latest addition to such Turn On sensors, allowing easy way detection and quantification of biothiols in living cells [bib_ref] Rapid turn on ESIPT fluorescent probe for colorimetric and ratiometric detection of..., Wang [/bib_ref].
For site specific protein labeling in live cells, Roger Y. Tsien and colleagues in 1998 pursued the use of biarsenical reagents. In real sense, it is the high affinity interaction of arsenic for thi-ols that forms the basis of biarsenical labeling technology. The fluorescent derivative FlAsH contains two arsenic atoms at a set distance from each other. Similarly, ReAsH is modified to contain resorufin. With the aid of fluorescence microscopy, FlAsH and ReAsH technology has been exploited to bind to tetracysteine (TC) sequences. The common TC sequence employed for this purpose is the six aminoacid sequence Cys-Cys-Pro-Gly-Cys-Cys. When bound to ethane dithiol (EDT), both FlAsH and ReAsH are non-fluorescent. Thus, upon binding with recombinant proteins containing such a tetracysteine motif, both these biarsenical labeling reagents-FlAsH-EDT and ReAsH-EDT turn to become highly fluorescent with green or red color respectively, with subsequent displacement of EDT.
Another most important bacterial enzymatic protein Haloalkane dehalogenase a hydrolase is modified and designed to covalently bind to a synthetic ligand with subsequent fusion to the protein of interest. Therefore, this enzyme has been efficiently exploited for visualizing subcellular localization of protein of interest, to capture the binding partners of a protein or for protein immobilization [bib_ref] The fluorescent toolbox for assessing protein location and function, Giepmans [/bib_ref]. The function of such a single protein tag is altered by attaching different chemical moieties of synthetic ligands called HaloTag ligands, through a chloroalkane linker, attached to useful molecules such as fluorescent dyes, affinity switches or solid surfaces. The Halotag is a modified haloalkane dehalogenase designed to covalently bind to synthetic ligands (Halotag ligands). It needs to be understood that, covalent bond formation between FP tag and the chloroalkane linker is extremely specific, occurs swiftly under physiological settings, and is essentially irreversible. However, the choice of ligand is done in accordance with the type of experiments to be performed.
Because of the heavy load of metal pollutants, researchers have come up with some important electrochemical sensors to detect different metals in a specific, sensitive and selective way. For example, to detect attomolar (aM) concentrations of Hg 2+ a precious electrochemical sensor was developed [bib_ref] Electrochemical sensor based on electrodeposited graphene-au modified electrode and NanoAu carrier amplified..., Zhang [/bib_ref]. For detection of this target metal, three single-stranded DNA probes were rationally designed. These probes were developed due to combined thymine-Hg 2+ -thymine (T-Hg 2+ -T) based coordination chemistry principles. In a similar approach, a sensitive electrochemical lead ion sensor was developed for lead (Pb 2+ ) ion detection. It was observed that, it was due to Pb 2+ -induced G-rich DNA conformation that upon Pb 2+ addition, DNA duplex got unwound and formed a stabilized G-quadruplex (G4) [bib_ref] Highly sensitive electrochemical sensor using a MWCNTs/GNPs-modified electrode for lead (II) detection..., Zhu [/bib_ref]. Another approach for detecting Pb 2+ , was based on an electrochemical DNA sensing strategy through modification of a glassy carbon electrode, with ordered mesoporous carbon nitride, gold nanoparticles and methylene blue [bib_ref] Electrochemical DNA sensing strategy based on strengthening electronic conduction and a signal..., Zeng [/bib_ref].
## Quantum dots
Quantum dots (QDs) contain some hundred to thousand atoms within a size of nanometer scale, distinctively made of an element of silicon or germanium or composed of a core of CdSe or CdTe and a ZnS shell [fig_ref] Figure 2: Application of different types of fluorophores, exploited for labeling and detection of... [/fig_ref]. Such petite structures vary in their color. The sharp fluorescence of such nanocrystals at discrete wavelengths depends on their size. Compared to other fluorophores, QDs possess 10-100 times higher extinction coefficients and better quantum yields [bib_ref] Quantum dot bioconjugates for imaging, labelling and sensing, Medintz [/bib_ref]. Here, a lone excitation wavelength readily excites the QDs of numerous emission spectra. QDs are preferably developed with coatings for biological investigations to facilitate their solubility in aqueous medium, help in conjugation with antibodies and prevents quenching by water [bib_ref] Advances in fluorescence imaging with quantum dot bio-probes, Pinaud [/bib_ref]. However, QD conjugated biomolecules lack efficient migration potential through intact cellular membranes due to their larger size, and thus, their expenditure for endocytosed proteins or extracellular apartments, permeabilized cells is restricted. Remarkable about QDs is their tolerance to repetitive imaging of solitary molecules, by virtue of their photo-stability [bib_ref] Review-quantum dots and their application in lighting, displays, and biology, Frecker [/bib_ref]. The portable electron-density and size permits analogous electron microscopy to be used for visualization of different objects of interest. However, a major issue with QDs is their toxicity in biological applications, and construction of conjugated polymer dots has helped a lot in this way.
Concerning the use of QDs to label target biomolecules, there is no strict protocol. However to achieve an optimized labeling, QDs are first made to become water dispersible and are then attached to target biomolecules. Many interactions are exploited to achieve QD fluorophore labeling with target biomolecules. Some major linking methods include-covalent linkages, biotin-avidin interactions and poly-histidine tags. Compared to small organic fluorophores, many biomolecules are attached to a single QD [bib_ref] Multiplexed toxin analysis using four colors of quantum dot fluororeagents, Goldman [/bib_ref] which leads to problematic orientation [fig_ref] Figure 2: Application of different types of fluorophores, exploited for labeling and detection of... [/fig_ref]. On the other hand, because of the better cell and organelle permeability of aryl fluorosulphates, they have been employed for the development of protein-selective covalent probes. In this scenario, an environment sensitive fluorogenic probe, 1,3,4-oxadiazole has been designed to bind selectively to transthyretin (TTR). Addition of this fluorogenic probe to HEK293T cells allowed efficient binding and imaging of cellular organelles like mitochondria and endoplasmic reticulum, making it a new fluorescent tool for living cells [bib_ref] A fluorogenic aryl fluorosulfate for intraorganellar transthyretin imaging in living cells and..., Baranczak [/bib_ref]. Upon application of this fluorogenic probe in Caenorhabditis elegans, it successfully detected TTR in six macrophage like cells.
## Fluorescent proteins
In real sense, the breakthrough of green fluorescent protein (GFP) from Jelly fish (Aquorea victoria) and then subsequent engineering of various other novel FPs from diverse organisms armed us with fluorophores, possessing extraordinary uniqueness for live cell imaging. Single GFP based sensors or chimeric FRET based nanosensors generate visible fluorescence for microscopic imagery. FRET involves the transfer of non-radiative energy from donor fluorophore to acceptor fluorophore, in the presence of a ligand. Such novel approach has been utilized to develop chimeric proteins that serve as nanosensors. These FRET based nanosensors consist of cyan fluorescent protein (CFP), a ligand binding domain and a yellow fluorescent protein (YFP) [fig_ref] Figure 2: Application of different types of fluorophores, exploited for labeling and detection of... [/fig_ref] , with CFP and YFP as the two mutant forms of GFP. However, after GFP, large spectrums of FPs with varied colors were discovered from marine coelenterates. The increase in brightness and folding efficiencies, accompanied with a decrease in oligomerization is achieved by generation of affinity mutants. This leads to the diversification of spectral range of FPs, from one color to another, improving the overall protein monitoring system of the resultant biosensor. Mutagenesis increases both photo-stability and photo-switchability of FPs [bib_ref] Improving the photostability of bright monomeric orange and red fluorescent proteins, Shaner [/bib_ref] [bib_ref] Highly photostable, reversibly photoswitchable fluorescent protein with high contrast ratio for live-cell..., Zhang [/bib_ref]. It is exactly the reversibility and irreversibility in photo-switching that makes FRET based biosensors useful to track protein trafficking. Quenching enforced by acidic pH is the major snag in the bio-sensing mechanism of FPs. However, recent developments have now better engineered the sensitivity of biosensors vis a vis ions, pH and redox potentials [bib_ref] Creating new fluorescent probes for cell biology, Zhang [/bib_ref] [bib_ref] Engineering and characterization of a superfolder green fluorescent protein, Pédelacq [/bib_ref].
Prior to FRET based biosensors, phyco-bili proteins and cyanobacterial photosynthetic antenna pigments were used as prime tags for fluorescence. It is due to their bigger size that problems arise in their diffusion which renders them limited only for surface labeling purposes [bib_ref] Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons, Karas [/bib_ref]. Therefore, such FPs are routinely utilized in conjugation with antibodies for enzyme-linked immunosorbent assays and flow cytometric measurements. For the FPs to offer their best character as fluorophores, some vital features include-varying spectral properties, maturation efficiency, photo stability, efficient brightness, and fidelity in fusions, monomeric character and their potential efficiency as FRET donors or acceptors. Furthermore, critical mutations have been exploited to generate different FP variants with enhanced spectral properties-mVenus [bib_ref] A variant of yellow fluorescent protein with fast and efficient maturation for..., Nagai [/bib_ref] and mKO2 [bib_ref] Visualizing spatiotemporal dynamics of multicellular cell-cycle progression, Sakaue-Sawano [/bib_ref] are the two important examples of bright monomeric variants. To address spectral and structural snags of FPs, recently a family of GFP proteins in the cephalochordate Branchiostoma floridae, commonly known as amphioxus was discovered and named after it as bfloGFP [bib_ref] Spectral and structural comparison between bright and dim green fluorescent proteins in..., Bomati [/bib_ref]. So far, this animal is credited to be the only species representing the largest source of 16 FPs. A total of six clades of amphioxus possess these 16 GFPs, all emitting the green fluorescent light, and each clade owns discrete absorption spectra, extinction coefficients and fluorescence intensities.
Utilizing the x-ray crystallography derived three dimensional (3D) structures, biochemical and spectral characteristics of two FPs-a bright FP (bfloGFPa1) and a weak FP (bfloGFPc1), this group [bib_ref] Spectral and structural comparison between bright and dim green fluorescent proteins in..., Bomati [/bib_ref] deliberated about the role of structural differences in FPs vis a vis chromophore environments that modulate into their photonic properties. A latest approach [bib_ref] Improving brightness and photostability of green and red fluorescent proteins for live..., Bajar [/bib_ref] has been carried out to improve the brightness and photostability of green and red FPs for enhanced live cell imaging in FRET. The group reported an improved photostability of mClover3 a derivative of GFP and mRuby3 a derivative of Red Fluorescent Protein (RFP) by an extent of 60% and 200% respectively over the previous generation of fluorophores. Also mRuby3 was recorded to be 35% brighter than its previously engineered version-mRuby2. Till the present period so far, out of all Jelly fish GFP and coral RFP derivatives, expressed in various mammalian cell lines, mClover3 and mRuby3 offers the highest fluorescence signals.
## Protein labeling schemes
In order to obtain the visualization of the protein of interest, two tagging schemes are generally followed-fluorescently labeled antibodies, in which antibodies are made to specifically bind with the protein of interest and the intrinsic fluorescent signal, where FP is genetically linked with the protein of interest. Though, in comparison to active antibodies, tagging through FPs is prolonged, yet when matched to QDs, genetic tags serve as the least cytotoxic fluorophores inside any cellular environment. The gene of interest is first cloned and then transformed into suitable cells. Once inside the cell, the bitterness due to FPs may be amplified, since they have their own unique proteinous feature which can lead to functional disorder of the attached protein of interest. In this direction, a striking effort [bib_ref] A monomeric red fluorescent protein with low cytotoxicity, Shemiakina [/bib_ref] was made to genetically engineer one least cytotoxic red FP (FusionRed) for monitoring the protein dynamics in real time. In the chromophore surrounding region K69R and R203H mutations were transferred to mKate2.5, resulting in the dramatic pH stability of the protein. Earlier, mKate2.5 variant was yielded by inserting S132A, R164A, K182E and Y200N mutations along with the substituted C-terminus. In addition to pH stability, the final variant FusionRed exhibited photostability and maturation rate at par with other red-emitting fusions. Here we deliberate on the labeling schemes of immunofluorescent tags and genetic tags.
## Labelling through immunofluorophores
There are two basic immunofluorescence techniques-direct and indirect. In former case, a single antibody conjugates directly with the fluorescent dye, whereas two different antibodies are exploited in latter-primary that recognizes target biomolecule and secondary that binds with fluorescent dye. However, for immunocytochemistry, direct labeling is carried out using both monoclonal and polyclonal antibodies. Background staining encountered with the use of secondary antibodies is also eliminated. In tissue sections, direct immunofluorescent technique (DIF) is principally applied in the identification of antibodies and other inflammatory proteins, to diagnose disease groups like pemphigus, lupus erythematosus, etc. that are histologically similar to separate under clinical investigations [bib_ref] Immunofluorescence and its application in dermatopathology with oral manifestations: revisited, Babu [/bib_ref].
Additionally, labeled streptavidin binding (LSAB) and avidinbiotin complex (ABC) are the two common immuno-histochemical techniques that have been recently commercialized. In LSAB, primary antibodies are conjugated directly to fluorophores and detected by streptavidin [bib_ref] Selective loading of kinesin-powered molecular shuttles with protein cargo and its application..., Ramachandran [/bib_ref] and in order to increase the spectral diversity for multiprotein analysis, antibodies are injected directly into live cells. Avidin-biotin labeling constitutes another receptor-ligand pair that works best in the secretory compartment. It is extremely useful for in vitro and histological investigations, however, for unknown reasons, this fluorophore toolbox is rarely explored in live cell imaging. pH measurements of various compartments in terms of different pK a values has been carried out through chicken avidin-biotin fluorescein conjugates, recombinantly expressed in different secretory compartments [bib_ref] Fluorescence photooxidation with eosin: a method for high resolution immunolocalization and in..., Deerinck [/bib_ref] [bib_ref] Seventeen-colour flow cytometry: unravelling the immune system, Perfetto [/bib_ref]. This facilitates the study of mechanism of pH regulation in such secretory compartments. However, primarily the critical limitation includes the toxic nature of avidin-biotin fluorescein conjugate in cytosol and mitochondrial compartments or secondly avidin is saturated with biotin. G protein coupled receptors (GPCRs) regulate critical physiological functions through neurotransmitters, peptide hormones, etc. Therefore to visualize GPCRs in living or fixed cells, immunofluorescence is applied through two strategies. In one approach, through antibodies against the extracellular receptor regions, and in other case through an epitope tag [bib_ref] Illuminating the life of GPCRs, Böhme [/bib_ref]. Intracellularly sited receptors or receptor segments and epitope tags are recognized and visualized only after cell fixation and permeabilization . The basic problem with immunolabelling is the larger . In order to visualize GPCRs, how in non-permeabilized cells: (A) antibodies are used against extracellular receptor regions or N-terminal epitope tags. (B) In contrast, it's only after cell permeabilization, that the visualization of intracellularly located receptors/segments or C-terminal epitope tags takes place. size of immunofluorophores, interfering with multiprotein recognition. Furthermore, immunofluorescent techniques are limited to permeabilized cells, extracellular proteins, and intracellular proteins. Whenever a problem arises due to availability of low-grade antibodies, an epitope tag is applied to the target protein to express it in a recombinant manner.
## Labelling through genetic tags
The major tagging hindrances have been overcome through genetically encoded FPs, which ensure perfect covalent tagging of FPs with the protein of interest. Genetic engineering followed by transformation procedure ensures an easier and perfect delivery of exogenous DNA into cells as compared to chemical dyes. By successfully developing different fusion constructs, FPs describes temporal dynamics of different metabolic processes in real time. For example, substrates binding to fusion constructs containing GFP undergo ubiquitin-proteasome dependent proteolysis [bib_ref] Short-lived green fluorescent proteins for quantifying ubiquitin/proteasome-dependent proteolysis in living cells, Dantuma [/bib_ref]. FRET disruption by proteases emerged as a classical biophysical parameter for investigators searching for novel drugs against major lethal diseases such as AIDS and cancer. FRET based biosensors possess a protease-sensitive linker fused between a blue fluorescent protein (BFP) and green fluorescent protein (GFP). The disruption of FRET by proteolysis separate the donor and acceptor fluorophores and individual domains of metallothionein (MT) were generated by proteolysis of CdMT with substilin [bib_ref] A fluorescence resonance energy transfer sensor for the -domain of metallothionein, Hong [/bib_ref]. In exact sense the cleavage occurred between two lysine residues-30 and 31 in the hinge region, ultimately cleaving the polypeptide chain between two fluorophores and thus in comparison to the control sample, energy transfer of dual-labeled MT decreased more significantly.
Cellular metabolite quantification in live cells constitutes the most important application of fluorescent proteins. For example, by inserting the peptide linker CaM and M13 between CFP and YFP, cameleons-the genetically encoded Ca 2+ sensors were constructed [bib_ref] Creating new fluorescent probes for cell biology, Zhang [/bib_ref]. By increasing the intracellular levels of Ca 2+ the affinity of CaM for the adjacent M13 sequence gets switched on. This causes a change in distance between two FPs that ultimately results into a large FRET. By replacing glutamate with glutamine residues in the pockets of Ca 2+ −binding sites, the effective affinity of Ca 2+ for cameleons is regulated. Recently, well optimized FRET based nanosensors have been developed for measuring Ras and Rap1 activity [bib_ref] Spatio-temporal images of growth-factor-induced activation of Ras and Rap1, Mochizuki [/bib_ref] and imaging glutamate levels in brain [bib_ref] Imaging of glutamate in brain slices using FRET sensors, Dulla [/bib_ref]. We have successfully constructed special FRET based nanosensors for quantification of leucine [bib_ref] Genetically encoded FRET-based nanosensor for in vivo measurement of leucine, Mohsin [/bib_ref] and methionine [bib_ref] Genetically-encoded nanosensor for quantitative monitoring of methionine in bacterial and yeast cells, Mohsin [/bib_ref] , for in vivo monitoring of zinc concentrations [bib_ref] Genetically encoded FRET-based nanosensor for in vivo monitoring of zinc concentration in..., Mohsin [/bib_ref] , lysine flux [bib_ref] Designing, construction and characterization of genetically encoded FRET-based nanosensor for real time..., Ameen [/bib_ref] and glycine betaine levels [bib_ref] Live cell monitoring of glycine betaine by FRET-based genetically encoded nanosensor, Ahmad [/bib_ref] and vitamin B12 levels [bib_ref] Live cell imaging of vitamin B12 dynamics by genetically encoded fluorescent nanosensor, Ahmad [/bib_ref].
## Protein expression and localization studies
Fluorescent flow cytometry facilitates the visualization of endogenous proteins in their active state, very important for single cell profiling. However, immunofluorescent tagging is most suitable for endogenous proteins. To visualize the expression of particular proteins in situ within different cell lineages or individual cells, mutant phenotypes can be analyzed by immunofluorescence staining. Although due to ample brightness of QDs, the detection limit advances much more and hence multi-label separation is achieved by varying spectral features. Though at light microscopic (LM) and electron microscopic levels (EM), QDs serve as appropriate tools for detection of proteins, however, it is only with the aid of electron microscopy that protein localization is effectively accomplished in subcellular structures. QDs are promising in electron microscopic observations, due to their size and electron dense core, and additionally, visualization is enhanced by silver staining. Recently, QDs have been utilized directly for concurrent immunolabelling of several endogenous proteins to acquire a correlated LM-EM imagery of cells [bib_ref] Correlated light and electron microscopic imaging of multiple endogenous proteins using Quantum..., Giepmans [/bib_ref]. Before proceeding for EM examination studies, we can swiftly get a precise analysis in LM by utilizing diverse range of QD labels. The fluorescent QDs at EM level overcome the need of antibody labeling, for example streptavidin labeled QDs detect surface proteins by using enzyme biotin ligase [bib_ref] Targeting quantum dots to surface proteins in living cells with biotin ligase, Howarth [/bib_ref]. Correlated microscopy using tetracysteine ensures the preservation of ultrastructure against the immunolabelled microscopic imagery that needs permeabilization and makes inflexible fixation. Both FPs and immunofluorescent tags are extensively applied for subcellular allotment of proteins and there is a high correlation between live-cell and fixed localization investigations. Researchers utilize finest validated antibodies and suitable fixation procedures, critical to obtain better epitope accessibility and an accurate protein distribution in vivo. In terms of cellular and subcellular localization however, FRET based nanosensors are comparatively more efficient than all other labeling methods. Compared to adding chemical dyes exogenously, such nanosensors can be expressed within stable cell lines and transgenic animals. The strategic approach is combination of FRET with metabolite recognition capability of a bacterial periplasmic binding protein (PBP) [bib_ref] Visualization of maltose uptake in living yeast cells by fluorescent nanosensors, Fehr [/bib_ref] , followed by successful transfection of resultant nanosensor into any cell type. This expands the possibilities for high-throughput screening, rigorous cell, developmental and physiological studies. In case of plants, most cells contain a large vacuole in their cytosol making it difficult to quantify plant metabolite concentrations in a given compartment in the little available cytoplasmic space. The FRET based nanosensors are comfortably targeted to sub-cellular locations, facilitating the highresolution mapping of signals within plant cells. Since there is an enormous range of metabolites in plants with highly complicated transport pathways that are quite contrasting against the animal systems.
## Visualization of protein activity in live cells
## Monitoring protein conformational changes and protein-protein interactions
A healthier spatiotemporal resolution of protein conformational changes is usually achieved by fixing a protein domain in between two FPs-Cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP) respectively. This is followed by FRET between two FPs, as the ligand binds with ligand sensing domain. Protein that connects the two FPs is engineered to undergo conformational changes in reaction to very important signals. With the aid of an appropriate localization signal, the resultant biosensors are targeted to precise subcellular compartments. A wide range of FRET based nanosensors have been constructed for measuring various metabolites, to monitor the activity of many proteases, check the balance between kinases and phosphatases, for sensing neurotransmitters and other metabolites. FRET efficiency varies with distance and orientation between donor and acceptor FPs. However, circular permutation of any of the two FPs [bib_ref] Expanded dynamic range of fluorescent indicators for Ca2+ by circularly permuted yellow..., Nagai [/bib_ref] or slight adjustments in the length of linker region greatly alters the FRET, as a result the frequent crop up of FRET responses are more due to reorientation and less by means of change in their inter fluorophore distance.
In living cells, FRET visualizes the dynamic protein-protein interaction, by means of two FP tagged proteins, whence a ligand binds with the ligand binding domain [fig_ref] Figure 4: FRET phenomenon [/fig_ref]. FRET processing occurs as the two FPs are within the intermolecular distance of 6-8 nm. However, recent approaches extended to the three FP tagged FRET based biosensors. It was carried out by valuable addition of monomeric red fluorescent protein (mRFP) to the pair of CFP/YFP. Such trimeric FRET based nanosensors again endow with deeper resolutions. In such FRET experiments, CFP acts as a donor for YFP and in turn, YFP as the next donor for mRFP [fig_ref] Figure 5: Three fluorophore based FRET process [/fig_ref]. On the basis of same principle, bimolecular fluorescence complementation (BiFC) is low pace chromophoric interaction between two split fragments of a FP [fig_ref] Figure 6: BiFC leading to the formation of a functional chromophore [/fig_ref] fused with two interacting proteins. Fluorescence correlation spectroscopy works in a novel way-using the two different tags aided with two different colors that interact and diffuse as a pair [bib_ref] Single-molecule spectroscopic methods, Haustein [/bib_ref]. There are simple colocalization events that indicate substantial protein-protein interactions. As seen in case of protein kinase A [fig_ref] Figure 7: Indirect FRET [/fig_ref] , its regulatory subunit and catalytic subunit are co-expressed. The latter is fluorescently tagged and destined to plasma membrane, and whence the adenosine 3 , 5 −monophosphate (cAMP) levels are increased, the two partners get dissociated, and therefore FRET visualization takes place.
Halotags have been used for FRET technique and it has been validated that compared to other bio-conjugated dyes, utilizing the Halotag protein linked with enhanced green fluorescent protein (GFP) displayed superior fluorescence stability [bib_ref] Quantitative comparison of different fluorescent protein couples for fast FRET-FLIM acquisition, Padilla-Parra [/bib_ref]. Halotags have been applied to monitor protein-protein and protein-DNA interactions. They have been used to monitor the interaction of bromodomain protein (BRD4) and histone deacetylase (HDAC1) along with other additional proteins [bib_ref] Discovering protein interactions and characterizing protein function using HaloTag technology, Daniels [/bib_ref]. Halotag has been adapted for the investigation of epidermal growth factor receptor Rasextracellular signal regulated kinase (ERK) mitogen activated protein (MAP) kinase pathway in living cells. FRET has been further adapted for Halotag mediated conjugation and site specific decoration of molecular beacons, utilizing two different FP fusions, thus enabling easy detection of target nucleotide sequences [bib_ref] Halo-tag mediated self-labeling of fluorescent proteins to molecular beacons for nucleic acid..., Blackstock [/bib_ref]. In live cells, the protein-protein interactions and alternative protein conformations have been detected through bipartite tetracysteine display. Such profluorescent biarsenical reagents-FlAsH-EDT2 and ReAsH-EDT2 have led to detection of early protein misfolding events associated with Alzheimer's and Parkinson's diseases and for high-throughput screening of compounds that stabilize discrete protein folds [bib_ref] Surveying polypeptide and protein domain conformation and association with FlAsH and ReAsH, Luedtke [/bib_ref].
## Visualizing protein dynamics
Protein labeling enables us to track its redistribution within a cell by using various imaging techniques which facilitate visuals of protein dynamics. Imaging readouts are therefore sought for such translocation processes that includes-monitoring buildup of polyphosphoinositides through FP-tagging of pleckistrin homology domains, proteins diffusing at steady state rates or involved in inter-compartmental exchange. Intensity fluctuations are statistically interpreted in fluorescence correlation spectroscopy. These fluctuations are because of fluorescent object mobilities, focused in front of a laser. Correlation of such images gained through spectroscopy efficiently measures these fluorescent variations. This facilitates the easy way illustration of mobility and interfacings of labeled molecules in real time. The critical factor for photomarking method is photochemical sensitivity. It involves dequenching of fluorophores, either through photoactivation [bib_ref] The uses of green fluorescent protein in mammalian cells, Lippincott-Schwartz [/bib_ref] or FP destruction. These two photo-marking modifications in fluorophores lead to the imaging of participant proteins. However, it needs to be emphasized, that the role of large fluorescent tag sizes is critical for passive mode of translocation.
This technique sees the tracking down of unimolecular protein translocations. It harvests the dual only fluorescent behavior (both poor and bright fluorescence in molecules or organelles) and uses automated software to track visualizations in the form of single videos. In the macromolecular structures of filamentous protein aggregates like actins and microtubules, upon fluorescent tagging the whole protein turnover processing is echoed by dynamic translocation of fluorescent patches. QDs were conjugated with EGF that resulted in specific co-localization with ErbB1-GFP chimeric receptor [bib_ref] Quantum dot ligands provide new insights into erbB/HER receptor-mediated signal transduction, Lidke [/bib_ref]. It was revealed that against activated ErbB1, Cy5-conjugated antibodies determined the activation of the receptor. Colocalization of transferrin labeled with Alexa fluor revealed that the complex internalization was a clathrin dependent phe- . Showing the process of photoactivation, in which intense illumination causes the quenching of a fluorescent protein, changing its spectrum. This is followed by movement of fluorophores towards each other. nomenon. By utilizing, distinctly synthesized EGF-QD, it was found that receptor oligomerization lead to retrograde transportation of receptor bounded single EGF-QD. It was furthermore revealed by the novel phenomenon of fluorescent recovery after photo bleaching (FRAP) that both actin flow and retrograde receptor transport occurred simultaneously [fig_ref] Figure 9: Depicting the intense illumination causes bleaching of the fluorescent protein in case... [/fig_ref]. With the use of biotin Alexa, the specific quenching of EGF-QDs revealed that receptor internalization occurred at cell body and not on filopodia. Henceforth, the involvement of filopodia to transport the activated receptors to cell body is brightly persuasive.
In the latest approaches an RFP based cAMP indicator "Pink Flamindo" applicable in optogenetics and in vivo imaging was developed. It was observed that, in presence of a saturating cAMP dose, fluorescence intensity of Pink Flamindo increased 4.2-fold at 567 nm and 590 nm excitation and emission peaks respectively. Recently, a FRET based system for detection of oversulfated chondroitin sulfate was constructed using the energy donor supercharged green fluorescent protein and dye labeled Hep (Hep-RF1) as the energy acceptor. Upon giving heparinase treatment to this sensor system, Hep-RF1 got hydrolysed into minute fragments, thus quenching the FRET signal [bib_ref] Red fluorescent protein-based cAMP indicator applicable to optogenetics and in vivo imaging, Harada [/bib_ref].
## Tracking protein synthesis and yield
It is imperative to gather the important information about protein synthesis and yield, involved in diverse physiological and developmental processes of a cell. However, in case of living cells or tissues, in no way we can adopt the methodologies that incorporate radioactive amino acids into nascent polypeptides. To solve this problem, only genetic tagging offers a great scope. The molecular strategy for tracking down the protein synthesis and yield involves the labeling of proteins with fluorescent probe of one color at initial stage of process up to a time limit. A new fluorescent probe of different color is used for labeling after an adequate time delay. For chimeric proteins, the phenomenon of bleaching and photo activation is explored to make the synthesized proteins get visualized. Varying temperature gradients lead to irreversible misfolding of FP mutants. Some FPs mature and change color spontaneously from green FP to red FP over a period of hours. Commonly, flexible linkers without hydrophobic residues are applied for monitoring the protein yield. Though in some cases, compared to flexible linkers, helix-forming hydrophobic linkers have been found to be better [bib_ref] A guide to choosing fluorescent proteins, Shaner [/bib_ref] , GFP tags are however removed through an enzymatic cleavage site (e.g., enterokinase) already engineered into the linker. Splicing of GFP or linker to N terminus of the target protein is usually preferred if it does not impairs its function and stability. In a novel investigation [bib_ref] Monitoring protein synthesis in living cells with fluorescent labeled tRNA FRET pairs, Barhoom [/bib_ref] protein synthesis has been monitored by exploiting fluorescent labeled tRNA FRET pairs in live cells through protein monitoring system (PSM). The synthesis of a viral protein during infectious stage was monitored through PSM using isoleucine tRNA. In a similar fashion, in mouse, synthesis of collagen in fibroblasts was monitored using tRNA-Gly and tRNA-Pro.
In chromophore activated light inactivation (CALI), protein activity is manipulated by immuno-targetting the protein of interest with organic dye [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref] , as observed in malachite green [bib_ref] Chromophore-assisted laser inactivation (CALI) to elucidate cellular mechanisms of cancer, Jay [/bib_ref]. Different fluorophore targeting has been investigated to overcome [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref]. Chromophore activated light inactivation: As a result of intense illumination of fluorescent dye, the target protein gets oxidized and inactivated due to ROS generation. This results in the loss of its fluorescent property. [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref]. Photo-oxidation is carried out for electron microscopy. The generation of ROS due to intense illumination causes the precipitation. Staining is carried out by osmium tetraoxide for imaging it on an electron microscope. the post delivery problems of dye conjugated antibodies into living cells. Mostly, the tetracycline bound biarsenical dyes ReAsH [bib_ref] Genetically targeted chromophore-assisted light inactivation, Tour [/bib_ref] and FlAsH [bib_ref] Transgenically encoded protein photoinactivation (FlAsH-FALI): acute inactivation of synaptotagmin I, Marek [/bib_ref] , genetic tags GFP [bib_ref] Dissecting the link between stress fibres and focal adhesions by CALI with..., Rajfur [/bib_ref] and other fluorescein conjugates are employed in molecular works to achieve the obliteration of proteins expressed ectopically. Another novel approach to minimize non-specific toxic effects involves shifting towards longer wavelengths, that are least damaging. The obligate dimer is the genetically encoded prototype for CALI known as KillerRed. Though being less efficient than ReAsH, it operates as an effective toolkit against photo-oxidation [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref] of FPs [bib_ref] A genetically encoded photosensitizer, Bulina [/bib_ref]. FRET process has been further exploited in tune with FlAsH and ReAsH based biarsenical dyes. In this scenario, fluorescence of a donor fluorophore attached to a protein is measured. This is followed by reaction of biarsenical acceptor dye to tetracysteine motif and the quantity of donor quenching is measured, providing a direct measurement of FRET efficiency. In membrane receptors, this technique has been applied to draw structural transitions in membrane receptors.
## Visualizing enzymatic catalysis
Many fluorescent imaging techniques have been explored to monitor and visualize activity of proteases, considering their biochemical importance in tumorigenesis, apoptosis, metastasis and inflammatory responses. Usually, the FRET disruption has been done with the use of proteases like subtilisin (Hong and Maret [bib_ref] A fluorescence resonance energy transfer sensor for the -domain of metallothionein, Hong [/bib_ref]. A comparative analysis of subtilisin treated metallothionein and those of free FPs remarkably abolished transfer of energy from donor FP to acceptor FP. In order to get a valuable understanding of proteolysis w.r.t. different inhibitors, good advances have been made to develop a range of fluorescence based assays. This facilitates us to unravel protease activities working in cells. For example, inhibitors of unrelated serine proteases influence the ruining of GFP based substrates in ubiquitin-protease system [bib_ref] Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM), Rust [/bib_ref]. Recently, it has been seen that GFP-reporter mice allows the in vivo tracking of proteasome activities, along with the outcome of proteasome inhibitors [bib_ref] A transgenic mouse model of the ubiquitin/proteasome system, Lindsten [/bib_ref]. Numerous works are also focusing to develop protease inhibiting compounds of life threatening viruses, e.g., hepatitis C virus [bib_ref] An NS3 protease inhibitor with antiviral effects in humans infected with hepatitis..., Lamarre [/bib_ref] and human corona virus, responsible for acute respiratory syndrome [bib_ref] Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs, Anand [/bib_ref]. In this direction, development of HIV-1 protease inhibitors was a major breakthrough to target viral proteases. Halotag technique has been tailored to study several disease models of bacteriology and virology. For example, Liu et al. studied the membrane topology of glycoprotein-41 of HIV in mammalian cells using two distinct Halotag ligands. In another study [bib_ref] Generation of a dual-functional split-reporter protein for monitoring membrane fusion using self-associating..., Ishikawa [/bib_ref] involving relative intensity of fluorescence from the Halotag and GFP system it was revealed that Halotag showed superior fluorescence intensity and also functioned better than GFP under acidic conditions.
Similarly, for caspases, which are cysteine proteases involved in apoptosis, numerous FRET based probes are available for monitoring their activity [bib_ref] Biochemical pathways of caspase activation during apoptosis, Budihardjo [/bib_ref]. Because, involvement of apoptosis within disease processing of neurodegenerative disorders [bib_ref] Biochemical pathways of caspase activation during apoptosis, Budihardjo [/bib_ref] and cancer [bib_ref] Apoptosis: a link between cancer genetics and chemotherapy, Johnstone [/bib_ref] , caspases are potential drug targets, considering their critical role as inducers and executors of cell death [bib_ref] Apoptosis-regulating proteins as targets for drug discovery, Reed [/bib_ref].The first functional caspase FRET [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref] was developed by fusing BFP, a peptide target sequence for caspase-3 effector and GFP [bib_ref] Detection of programmed cell death using fluorescence energy transfer, Xu [/bib_ref]. Due to the caspase-3 activation, as the apoptosis got induced, transfected cells expressing this chimeric protein lost their FRET signal between donor (BFP) and acceptor (GFP). However, as the BFP donor in BFP-GFP FRET pair is excited by ultraviolet light, it resulted in enough cellular damage, preventing its application as a fluorescent toolkit in living cells. This problem has been overcome by generating new spectral variants of FPs.
## Spatiotemporal resolution
In the myriad of cellular and molecular processes, operating over a wide range of distances (1-100 nm), efforts are being done to overcome the microscopic limits (>200 nm) and achieve a definite spatiotemporal resolution. The major advantage of this approach facilitates the dimensional interpretation of motor proteins and other downstream enzymes [bib_ref] Myosin V walks hand-over-hand: single fluorophore imaging with 1.5-nm localization, Yildiz [/bib_ref]. In recent approaches, to get a high resolution imaging up to 1 mm depths, infrared multiphoton excitation is very advantageous, however, this technique must fulfill the critical condition of collection and harnessing emission photons that have originated from illumination focus. Though tomography is favored for imaging of live tissues yet its major bottleneck is seen in reduced resolution and this problem is solved by serial reconstruction technique utilized to visualize samples through fluorescence imaging at greater depths [bib_ref] All-optical histology using ultrashort laser pulses, Tsai [/bib_ref].
Latest developments for better resolutions explain FPs have evolved the photo-activated localization microscopy (PALM). Point spread function (PSF) of a microscope represents the intensity outline of focused emission from a lone protein molecule producing a diffracted spot and a blurred spot. In case visible light and high numerical aperture (NA) lenses are used, this spot size is approximately 200 × 500 nm along lateral and axial dimensions. Since the individual images are merged, there is no distinction of fluorescing protein molecules in the distances smaller than the dimensions of this spot. When there is simultaneous fluorescing of FPs, PALM ensures their separation much greater than this diffraction-limited distance [fig_ref] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study... [/fig_ref] , thus individually localizing each FP. Applying a low intensity light, in a densely labeled structure there occurs stochastic activation of photo-controllable FPs. Therefore, an appropriate separation distance is achieved between fluorophores [bib_ref] Imaging intracellular fluorescent proteins at nanometer resolution, Betzig [/bib_ref] [bib_ref] Ultra-high resolution imaging by fluorescence photoactivation localization microscopy, Hess [/bib_ref]. These activated molecules are then imaged and localized before being deactivated either through perpetual photo-bleaching or by turning off the fluorescence. However in order to collect such molecular positions in millions before deactivation, this process is repeated many thousand times. This microscopic technique of super-resolution imaging based on single molecule can work with all shades of photo-controllable FPs, whether rsFPs, PAFPs or PSFPs. Practically, in single molecule super-resolution microscopy, two critical factors-the fineness of localization of individual molecules and their density of localization regulate the spatial resolution. The microscopic technique involves photo-manipulation of such protein molecules. This type of imaging technology uses both synthetic dyes [bib_ref] Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM), Rust [/bib_ref] as well as photo-controllable FPs and has been applied in total internal reflection [bib_ref] Imaging intracellular fluorescent proteins at nanometer resolution, Betzig [/bib_ref] and epifluorescence [bib_ref] Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM), Rust [/bib_ref] [bib_ref] Ultra-high resolution imaging by fluorescence photoactivation localization microscopy, Hess [/bib_ref].
## Conclusion and future outlook
Fluorescent microscopic imagery is an exquisitely efficient technology that has come to the aid of researchers due to coordinated expansion of targeting strategies, fluorescent probes, highly optimized microscopes and precise data analyzing softwares, which synchronously facilitated live cell and multiprotein imaging, single molecule detection and high throughput screening. Mutagenesis on the other hand expanded the spectral diversity of natural FPs, generating a valuable range of phenotypes, which are very much desirable as non-cytotoxic fluorescent tags. A library of reporters has been developed for different metabolites, biochemical processes and regulatory enzymes, and it marks certainly boundless achievement in systems biology. For visualizing protein of interest through live cell imaging, although genetic tags provide a remarkable advantage, attempts should be made by genetic engineers to explore other fluorescent probes that cause no irritation to the functioning of endogenous proteins. For example, in the technique of immunofluorescence the detection limit is decreased with the application of QDs that are stable and possess increased brightness and this advantage amplifies its prospects for multiprotein detection, correlated EM examinations and other ELISA and Western blot based in vitro assays. However, while designing such prized QDs efficient targeting and better penetration is very crucial to achieve flawless fluorescence visualizations. In recent advancements, because of their efficient photo-physical and biochemical properties photo-controllable FPs has emerged as the key tools for super-resolution microscopy.
Each of three discussed fluorescent probe is equipped with their unique advantages and limitations. With organic dyes, the pulselabeling procedures are much easier. Secondly, targeting motifs are small and, therefore, a very less opportunity remains for introduced sequences to disrupt gross folding and function of a labeled protein. However, with biarsenical dyes, there are three fundamental problems during labeling; the effect of tag causes disturbed protein localization and function. The background fluorescence level and the cellular toxicity due to ligands further deteriorate the fluorescence intensity measurements and imaging in eukaryotic living cells. The largest problem associated with quantum dots is their toxicity, since these QD nanomaterials are composed of heavy metals, which are potentially toxic during in vitro and in vivo imaging. QD toxicity depends on different parameters such as charge, size, shape, composition, redox reactivity, surface coating, solubility, photostability and exposure time. On the other hand, whether GFP or its variants are non-toxic to cells but some limitations does exist. First, it is the aggregation of FPs and secondly, exciting GFP for an extended time generates free radicals. It has been further established that GFP variants induce apoptosis, that indicates the possible reason of difficulty in establishing the stable cell lines expressing GFP [bib_ref] Is green fluorescent protein toxic to the living cells?, Liu [/bib_ref]. Therefore, FRET based nanosensors undergo a healthy expression under in vivo environments of prokaryotic model organisms such as Escherichia coli and Saccharomyces cerevisiae, and animal cell lines. In practice, several fluorescent probe based sensors have been patented for the specific detection of target ligands, drugs or metabolites. From the time of the first report of colorimetric detection of influenza virus through a polydiacetylene (PDA) film, researchers developed a great interest in their application as efficient sensor chips [bib_ref] A polydiacetylene-based fluorescent sensor chip, Kim [/bib_ref]. PDA-based fluorescent chemo-sensor systems are compatible with microarray technologies. In response to a range of environmental perturbations such as temperature, pH and ligand-receptor interactions, these chemosensors undergo a blue to red visible color change. The important FRET based sensors for apoptosis detection [bib_ref] FRET-based apoptosis detector, Google Patents, Myc [/bib_ref] to detect caspase-3, and proteolytic activity of matrix metalloproteinase 9 (MMP-9) with high spatiotemporal resolution [88] are some of the novel sensing systems available to us.
Fluorescent probe based measurements and imaging technology has a vast role to play ahead. In medicine and clinical research, application of fluorescent probes for biopsies and biochemical tests such as assays to analyze profiles of protein activity in patient cells is emerging very fast. Furthermore, fluorescence detection will be applicable to precisely investigate in individual patients the special modulatory effects of drugs on cellular signaling. Thus, fluorescence assays ensure a high-throughput drug screening by means of both functional assays in animal models or live cells and through protein microarrays. To achieve a better spatiotemporal resolution, photo-activated localization microscopy guarantees the individual localization of each FP by means of a low intensity light in a densely labeled structure attaining a suitable separation between photo-controllable FPs, much greater than the diffraction-limited distance. Therefore, in the inquiries of metabolomics, proteomics and systems biology, the sui-generis alliance of high molecular specificity, harmonized and nondestructive compatibility with prokaryotic or eukaryotic organisms and living cells, and certain spatiotemporal resolution based microscopic technology will prevail as pivotal.
[fig] Figure 1: Outline of fluorescent probes, their types, applications and different methods to study protein dynamics and localization within a cell or whole organism. [/fig]
[fig] Figure 2: Application of different types of fluorophores, exploited for labeling and detection of proteins. (A) A 3T3 cell from an adherent culture, immunolabelled with Alexa Fluor 488. (B) Quantum Dot with CdSe core and ZnS shell. (C) A QD conjugated with many antibodies, preventing its mobility. (D) A ligand sensing domain, inserted between CFP and YFP. [/fig]
[fig] Figure 4: FRET phenomenon: FRET takes place when ligand binds with the ligand sensing domain, through which CFP (donor) and YFP (acceptor) are in proximity. [/fig]
[fig] Figure 5: Three fluorophore based FRET process: by the attachment of gene 1 with gene 2, intermolecular FRET occurs first between CFP and YFP. Afterwards, gene 3 encoding RFP also gets attached, which acts as new acceptor fluorophore for another FRET. [/fig]
[fig] Figure 6: BiFC leading to the formation of a functional chromophore. Before this process, the two GFP halves are separate and BiFC leading to their proper folding resulting into the formation of a functional chromophore. [/fig]
[fig] Figure 7: Indirect FRET: Addition of a phosphate group to the substrate by Kinase enzyme leads to a conformational change that brings CFP and YFP close together, which ultimately results into FRET. [/fig]
[fig] Figure 9: Depicting the intense illumination causes bleaching of the fluorescent protein in case of FRAP. The fluorescence is however restored back by the diffusion of nonbleached fluorophore CFP. [/fig]
|
Exercise training improves peak oxygen consumption and haemodynamics in patients with severe pulmonary arterial hypertension and inoperable chronic thrombo-embolic pulmonary hypertension: a prospective, randomized, controlled trial
# Introduction
Previous studies in patients with idiopathic pulmonary arterial hypertension (IPAH), inoperable chronic thrombo-embolic pulmonary hypertension (CTEPH), [bib_ref] Exercise training improves exercise capacity and quality of life in patients with..., Nagel [/bib_ref] associated PAH (APAH) in connective tissue disease, [bib_ref] Exercise training in pulmonary arterial hypertension associated with connective tissue diseases, Grunig [/bib_ref] and APAH in congenital heart disease [bib_ref] Efficacy of exercise training in pulmonary arterial hypertension associated with congenital heart..., Becker-Grunig [/bib_ref] have shown beneficial effects of exercise training as add-on to disease-targeted medical therapy increasing exercise capacity and quality of life (QoL). [bib_ref] Exercise and respiratory training improve exercise capacity and quality of life in..., Mereles [/bib_ref] [bib_ref] Safety and efficacy of exercise training in various forms of pulmonary hypertension, Grü Nig [/bib_ref] [bib_ref] Benefits of intensive treadmill exercise training on cardiorespiratory function and quality of..., Chan [/bib_ref] In four smaller studies, a randomized, controlled trial design has been performed. [bib_ref] Exercise and respiratory training improve exercise capacity and quality of life in..., Mereles [/bib_ref] [bib_ref] Ambulatory rehabilitation improves exercise capacity in patients with pulmonary hypertension, Fox [/bib_ref] [bib_ref] Benefits of intensive treadmill exercise training on cardiorespiratory function and quality of..., Chan [/bib_ref] [bib_ref] Effect of aerobic exercise training on fatigue and physical activity in patients..., Weinstein [/bib_ref] These trials assessed patients with IPAH and inoperable CTEPH and showed an improvement of 6 min walking distance (6MWD), cardiorespiratory function, fatigue, activity, and QoL parameters by exercise training. In further prospective, uncontrolled trials, similar results have been obtained and long-term follow-up revealed excellent 1-and 2-year survival rates of 95 -100% in the patients who had participated in the training programme and continued training at home. The sample size of all these studies was relatively small . Despite these limitations, the recommendation for this thoroughly supervised rehabilitation including exercise training has been upgraded to Class I, with a Level of Evidence of A within the last world congress in Nice in 2013. [bib_ref] Updated treatment algorithm of pulmonary arterial hypertension, Galie [/bib_ref] [bib_ref] Clinical worsening in trials of pulmonary arterial hypertension: results and implications, Galie [/bib_ref] This recommendation was combined with the request to perform further randomized controlled multicentre trials. [bib_ref] Clinical worsening in trials of pulmonary arterial hypertension: results and implications, Galie [/bib_ref] The mechanisms for the improvement of symptoms, exercise, and functional capacity and the possible effects on prognosis are still not completely understood.
The improvement of muscle function such as increase in capillarization and change of fibre type by exercise training 2 has already been shown in PH. There is also evidence that exercise training influences the pulmonary vasculature, [bib_ref] Structural and functional prevention of hypoxia-induced pulmonary hypertension by individualized exercise training..., Weissmann [/bib_ref] with a regulating effect on pulmonary vascular remodelling. The effect of exercise training on peak oxygen consumption (VO 2 /kg) is inconsistent in PH trials. Few uncontrolled trials using exercise training in an outpatient setting did not observe a significant improvement in VO 2 . [bib_ref] Vonk-Noordegraaf A. Effects of exercise training in patients with idiopathic pulmonary arterial..., De Man [/bib_ref] [bib_ref] Rehabilitation program in adult congenital heart disease patients with pulmonary hypertension, Martinez-Quintana [/bib_ref] [bib_ref] Results of a 12-week outpatient cardiovascular rehabilitation in patients with idiopathic pulmonary..., Boutet [/bib_ref] Up to now, there is no study investigating the effect of exercise training on peak VO 2 /kg as primary endpoint, allowing a confirmative conclusion. Furthermore, there is growing evidence that training may improve right ventricular (RV) function. Animal models suggested a beneficial effect of training on the RV, such as a reduction of RV end-diastolic pressure and increase in RV capillary density (+86%, P , 0.05). [bib_ref] Effects of exercise on monocrotaline-induced changes in right heart function and pulmonary..., Colombo [/bib_ref] [bib_ref] Opposite effects of training in rats with stable and progressive pulmonary hypertension, Handoko [/bib_ref] The effect of exercise training on cardiopulmonary function and RV haemodynamics is of high interest and needs to be further assessed using cardiopulmonary exercise testing (CPET) and right heart catheterization (RHC) before and after training. Recent publications have shown that peak oxygen uptake (VO 2 ), cardiac index (CI), [bib_ref] Prognostic value of exercise pulmonary haemodynamics in pulmonary arterial hypertension, Chaouat [/bib_ref] and systolic pulmonary arterial pressure during exercise are important independent predictors for survival. Therefore, the aim of this study was to perform a large, prospective, randomized trial to investigate the impact of exercise training on peak VO 2 as an important prognostic factor in patients with severe PAH or CTEPH and right heart failure. The study was also aimed to provide first insights into the effects of exercise training on invasively measured haemodynamics.
# Methods
## Study population and design
Patients with PAH and inoperable or persistent CTEPH and chronic right heart failure who were stable on disease-targeted medication for at least 2 months prior to inclusion were randomly assigned to a control and a training group. Medication remained unchanged during the study period. Clinical examinations including RHC were performed at baseline and after 15 weeks. During the study, patients of the control group did not receive any advice on exercise training [fig_ref] Figure 1: Study design [/fig_ref].
All patients gave their written informed consent to the study. The study was approved by the Ethics Committee of the University of Heidelberg, Germany. The study was registered at clinicaltrials.gov (NCT01394367).
## Outcome measures
Primary endpoint was the change in peak VO 2 /kg. Secondary endpoints were the change from baseline to 15 weeks in invasively measured haemodynamics at rest and during symptom-limited exercise, 6MWD, workload QoL questionnaire (SF-36), World Health Organization (WHO) functional class, and N-terminal prohormone brain natriuretic peptide (NT-proBNP). QoL questionnaires were analysed in a blinded fashion. Patients in the control group went on with their usual lifestyle during the study. Efficacy parameters were evaluated at baseline and week 15, as described previously. 1,10 Assessment of 6MWD, SF-36, and other efficacy parameters were performed by investigators who were blinded to the clinical data. Changes in WHO functional class and Borg dyspnoea scale (with 6 representing no exertion and 20 maximal exertion) [bib_ref] Psychophysical bases of perceived exertion, Borg [/bib_ref] were also analysed. CPET was carried out as supine cycle exercise, as described previously. [bib_ref] Exercise and respiratory training improve exercise capacity and quality of life in..., Mereles [/bib_ref] Systolic and diastolic systemic blood pressures, workload, heart rate, ventilation, and VO 2 were measured breath-by-breath. In patients dependent on long-term oxygen therapy, oxygen supplementation was paused throughout and resupplied directly after termination of the test.
RHC was performed at baseline and 15 weeks. Patients were examined on a variable load supine cycle ergometer (model 8420; KHL Corp., Kirkland, WA, USA). The examination at rest and during symptom-limited exercise was performed in a supine position using the transjugular approach with an 8F introducer set (MXI100, MEDEX, Smiths Group Plc, UK). RHC was performed by triple-lumen 7F-Swan-Ganz thermodilution catheters (Edwards Lifesciences, Irvine, CA, USA). Workload started at 25 W and was increased every 2 min for 25 W until symptom-limited exercise capacity was reached. Cardiac output (CO) was measured at least in triplicate at rest and in duplicate at the end of each workload during exercise by thermodilution with a variation of ,10% between the measured values. The zero reference point for pressure recordings was set at the level of the right atrium in the mid-axillary line (phlebostatic axis). All examinations and measurements were performed by the same experienced team. There were no complications.
## Rehabilitation programme with exercise training
The exercise and respiratory training was performed as described before 1,10 and was started in-hospital for 3 weeks in the Rehabilitation Clinic Kö nigstuhl in Heidelberg. We performed a programme especially developed for PH patients with at least 1.5 h/day exercise training (in intervals distributed over the day), consisting of interval cycle ergometer training at low workloads at 7 days a week, walking, dumbbell training of single muscle groups using low weights (500 -1000 g), and respiratory training at 5 days/week. The training was continued at home with at least 15 min/day at 5 days a week for the following 12 weeks.
Besides physical training, patients received mental training to improve their perception of their individual physical abilities and limits. Psychological support was offered to all participants. The training programme was closely supervised by physical therapists and physicians specialized in rehabilitation medicine and by PH experts, as described before. [bib_ref] Exercise and respiratory training improve exercise capacity and quality of life in..., Mereles [/bib_ref] Adverse events were recorded whenever they occurred. Oxygen saturation and heart rate were monitored continuously throughout the training and were used to adjust the training intensity. When patients' oxygen saturation fell below 90% during exercise, they received oxygen supply (3 -10 L/min) throughout the training. Patients who were on oxygen therapy 16 -24 h/day before inclusion into this study remained on oxygen throughout the training programme. At discharge from hospital after 3 weeks, patients received an individualized training manual and ordered a cycle ergometer for use at home.
# Statistical methods
The analyses were performed by two statisticians (N.E. and C.F.). Data are given as mean values + standard deviations. For the description of effects, we present changes in absolute values and changes in per cent of the individual baseline value.
Calculation of sample size was based on the primary efficacy endpoint, which was defined as the change in peak VO 2 /kg between baseline and 15 weeks of exercise training. The null hypothesis to be assessed in the confirmatory analysis states that there is no difference in the changes of the peak VO 2 /kg value between the intervention and the control group. To detect a placebo-adjusted difference in peak VO 2 /kg of 2 mL/min/kg, with a standard deviation of 3 mL/min/kg, with a power of 80%, and a two-sided significance level of 5%, it was calculated that 45 patients were required in each treatment arm. Primary efficacy analysis was performed by an analysis of covariance (ANCOVA) model including baseline values as covariate, thereby yielding a power advantage over the standard t-test used for sample size calculation. Training improves cardiorespiratory function in PAH and CTEPH The secondary endpoints were compared between groups by an ANCOVA including baseline values as covariate. For analysis of categorical data, the McNemar -Bowden test was used.
To include missing values, we performed two sensitivity analyses: (i) multiple imputation analysis based on multiple regression models with 20 imputed data sets and (ii) a responder analysis regarding missing values and worsening of the parameters as non-responder. For the multiple imputation analysis, the combined results are stated. Threshold between response and non-response was set as 0. The responder analysis with number/percentage of responders vs. non-responders was performed by x 2 test. All tests were two-sided, and P-values less than 0.05 were considered statistically significant. All analyses were carried out with IBM SPSS V20 (IBM Corp. Armonk, NY, USA).
# Statement of responsibility
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
# Results
## Study population and randomization
Ninety-five consecutive patients with invasively diagnosed PAH and inoperable or persistent CTEPH under optimized medication therapy were included [fig_ref] Figure 1: Study design [/fig_ref]. Eight patients were not eligible for the study: one patient had an alteration in his/her medication, two patients were not able to perform an ergometer test, and five patients had an invalid ergometer test at baseline. Thus, the study group consisted of 87 patients who were randomized into a control and a training group (71% PAH, 29% CTEPH, 79% WHO functional class III, 54% female, 56 + 15 years) [fig_ref] Figure 1: Study design [/fig_ref] and [fig_ref] Table 1: Baseline characteristics [/fig_ref].
The control and training groups did not differ in their baseline characteristics, disease severity, or medication [fig_ref] Table 1: Baseline characteristics [/fig_ref].
Primary endpoint: change in peak oxygen uptake Peak VO 2 /kg significantly improved in the training group with +3.1 + 2.7 mL/min/kg (baseline 13.3 + 3.6 mL/min/kg; mean increase relative to baseline +24.3%) vs. 20.2 + 2.3 mL/min/kg (baseline 12.7 + 4.0 mL/min/kg; mean increase relative to baseline +0.9%) in the control group (P , 0.001; [fig_ref] Figure 2: Primary endpoint [/fig_ref] and [fig_ref] Table 2: Change echocardiography, CPET, and haemodynamics [/fig_ref]. In the training group, 91.7% of the patients reached a peak VO 2 /kg of ≥11.5 mL/min/kg, compared with the control group with 50%.
## Secondary endpoints
Haemodynamics and right heart function RHC was performed at baseline and after 17 + 7. [fig_ref] Figure 3: Secondary endpoints [/fig_ref].
Measurements during exercise baseline vs. 15 weeks revealed a significant improvement of CI during exercise (training +1.0 + 1.4 L/min/m 2 ; +19.5% vs. control 20.2 + 0.6 L/min/m 2 ; 24.3%, P ¼ 0.002; and further haemodynamic parameters in the training group, compared with the control group [fig_ref] Figure 3: Secondary endpoints [/fig_ref] and [fig_ref] Table 2: Change echocardiography, CPET, and haemodynamics [/fig_ref]. Both time points were also analysed for the same workload, which corresponds either to the maximum of baseline or follow-up examination. The analysis at the same workload showed the same results with significant improvement of both CI and CO during exercise (difference from baseline: CO training +1.41 + 2.13 L/min vs. control 20.38 + 0.29 L/min, CI training +0.66 + 1.06 L/min/m 2 vs. control 20.23 + 0.66 L/min/m 2 , both P ¼ 0.001 for original data; for multiple imputation: both P ¼ 0.03 compared with baseline and control).
## Further clinical parameters
After 15 weeks of treatment, the 6MWD and change in the workload during CPET significantly improved in the training compared with the control group (control group corrected change: 6MWD +41 m, P ¼ 0.001 and CPET +19 W, P , 0.001; and [fig_ref] Table 2: Change echocardiography, CPET, and haemodynamics [/fig_ref].
At baseline, all patients had significantly compromised QoL scores compared with the general population. Only for the subscale vitality, P-values were significant for both original data and multiple imputation (for original data P ¼ 0.036 and for multiple imputation P ¼ 0.042) [fig_ref] Figure 6: QoL [/fig_ref].
For evaluation of WHO functional class, the prepared sheets were not filled in by the investigators for .50% of the patients, so this parameter was not analysed. NT-proBNP values varied largely within the population at baseline also due to comorbidities such as renal insufficiency and did not show a statistically significant difference between groups (training 979 + 2811 vs. control 1074 + 1424; P ¼ 0.44). Echocardiography data showed no statistically significant change in right heart areas and systolic pulmonary arterial pressure between groups. Tricuspid annular plane systolic excursion significantly decreased in the training group compared with the control group.
## Sensitivity analysis: responder vs. non-responder
A sensitivity analysis was performed for primary and secondary endpoints treating missing values as non-responders. Change in workload (training 54% vs. control 21% responders, P ¼ 0.002), peak VO 2 (training 74% vs. control 44%, P ¼ 0.006), mPAP (training 64% vs. control 28%, P ¼ 0.001), CO (training 59% vs. control 23%, P ¼ 0.001), CI (training 51% vs. control 21%, P ¼ 0.005), PVR (training 64% vs. control 26%, P ¼ 0.001), CO during exercise (training 59% vs. control 21%, P ¼ 0.001), and CI during exercise (training 56% vs. control 23%, P ¼ 0.003) remained statistically significant.
# Discussion
This large prospective, randomized, controlled trial evaluates the effect of exercise training on peak VO 2 /kg as primary endpoint and invasively measured haemodynamics in patients with severe PAH and inoperable or persistent CTEPH. To the best of our knowledge, there is only one previous publication performing RHC after training in 20 patients with left heart failure. [bib_ref] Exercise training in patients with severe chronic heart failure: impact on left..., Erbs [/bib_ref] The results of the study showed an improvement of peak VO 2 /kg, CI, and CO during exercise, suggesting that this treatment might improve RV function.
Other prognostic parameters such as CI at rest, pulmonary vascular resistance, mPAP, 6MWD, QoL parameters, and exercise capacity significantly improved in the training group. Bearing in mind that these effects were reached as add-on in patients with optimized medical treatment, low-dose exercise training was very effective.
## Effect on peak oxygen uptake and rv function
The results of this study confirm previous findings that exercise training improves peak VO 2 /kg in patients with severe PH by approximately 15-25%. [bib_ref] Exercise and respiratory training improve exercise capacity and quality of life in..., Mereles [/bib_ref] [bib_ref] Safety and efficacy of exercise training in various forms of pulmonary hypertension, Grü Nig [/bib_ref] [bib_ref] Exercise training in pulmonary arterial hypertension associated with connective tissue diseases, Grunig [/bib_ref] [bib_ref] Efficacy of exercise training in pulmonary arterial hypertension associated with congenital heart..., Becker-Grunig [/bib_ref] [bib_ref] Effect of exercise and respiratory training on clinical progression and survival in..., Grünig [/bib_ref] Study results that did not show positive effects of exercise training on peak VO 2 /kg 2,14,15 may be evoked by less intensive training due to an outpatient setting and by a lower training frequency of three times a week or less. Peak VO 2 is linearly associated with RV function. [bib_ref] Pulmonary vascular hemodynamic response to exercise in cardiopulmonary diseases, Lewis [/bib_ref] [bib_ref] Exercise stress echocardiography of the pulmonary circulation: limits of normal and sex..., Argiento [/bib_ref] Both were severely shown on the left side of each category. Exercise and respiratory training significantly improved the subscale score for vitality (P ¼ 0.036), compared with the control group, in which this parameter slightly worsened. The subscales role limitations due to physical restrictions (P ¼ 0.099), role limitations due to emotional restrictions (P ¼ 0.09), and general health perception (0.091) were significant in trend in the original data, but showed inconsistent findings in multiple imputation. reduced in our patients at rest and during exercise. RV dysfunction is a crucial factor contributing to functional impairment and mortality. Improvement of peak VO 2 /kg is possibly due to improvement of capillary density of the skeletal muscle, which has been identified before in patients with IPAH. [bib_ref] Vonk-Noordegraaf A. Effects of exercise training in patients with idiopathic pulmonary arterial..., De Man [/bib_ref] CI and CO significantly improved at rest, during maximal exercise, and when comparing the baseline and follow-up assessments at the same workload. The improvements in haemodynamics at rest and during exercise point out that training may also be beneficial for the RV function, which is a strong prognostic predictor in these patients. [bib_ref] Assessment and prognostic relevance of right ventricular contractile reserve in patients with..., Grunig [/bib_ref] [bib_ref] Impact of right ventricular reserve on exercise capacity and survival in patients..., Blumberg [/bib_ref] [bib_ref] Right ventricular performance and contractile reserve in patients with severe heart failure...., Gorcsan [/bib_ref] [bib_ref] Prognosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines, Mclaughlin [/bib_ref] CI during exercise has recently been detected to be the most important independent haemodynamic predictor of survival in PH in the investigated cohort, 18,20 correlated well with the clinical data, and was even more informative than haemodynamics at rest.
## Mechanism of exercise training in athletes and patients with left heart failure
Previous studies in healthy athletes have shown that endurance exercise training causes a significant increase in skeletal muscle capillarization, characterized by an elevated capillary density and capillary-to-fibre ratio. The physiological adaptation contributes to enhanced aerobic capacity by increased transport, conductance, and extraction of oxygen in skeletal muscles. In patients with left heart failure, exercise training has shown to improve ventilatory efficiency, 31 the reversal of skeletal muscle atrophy, [bib_ref] Alterations of skeletal muscle in chronic heart failure, Drexler [/bib_ref] the attenuation of endothelial dysfunction, [bib_ref] Regular physical exercise corrects endothelial dysfunction and improves exercise capacity in patients..., Hambrecht [/bib_ref] and inflammatory mediators. Besides these 'peripheral effects' on skeletal muscle function, it is not known yet whether there is a direct effect of exercise training on the heart muscles. Erbs et al. [bib_ref] Exercise training in patients with advanced chronic heart failure (NYHA IIIb) promotes..., Erbs [/bib_ref] showed improvement of left ventricular ejection fraction, stroke volume, and CO at peak exercise after 12 weeks of exercise training in patients with advanced chronic left heart failure. [bib_ref] Pulmonary vascular hemodynamic response to exercise in cardiopulmonary diseases, Lewis [/bib_ref] Wisloff et al. [bib_ref] Superior cardiovascular effect of aerobic interval training versus moderate continuous training in..., Wisloff [/bib_ref] reported that after 12 weeks of aerobic-moderate continuous training and aerobic-intense interval training, left ventricular ejection fraction of cardiac patients increased. This study confirmed the effects of exercise training by invasive measurements of pulmonary circulation in patients with PAH/CTEPH and right heart failure.
# Limitations
RHC was an optional assessment in this study. Thus, a referral bias may have occurred as only 83% of the patients received RHC. The haemodynamic effects of this study may therefore serve as first results to plan prospective, randomized studies on the effects of exercise training on haemodynamics at rest and during exercise. The training group showed a significant increase in heart rate during exercise when compared with the control group. This may be a hint for a higher volitional effort. However, haemodynamics also showed a significant improvement when examinations were compared for the same workload.
In a randomized controlled trial, a responder analysis does not reflect a worst-case scenario and is therefore not appropriate as a mere sensitivity analysis. The findings of both responder analysis and multiple imputations, however, showed almost the same significant parameters as the original data analysis.
NT-proBNP values, which are associated with disease progression, showed no statistical difference between changes of control and training groups. This may also be influenced by the large variation of values, as in six patients NT-proBNP was highly increased due to renal insufficiency.
It is a general issue of exercise training studies that they cannot be performed in a blinded fashion and that a referral bias towards highly motivated patients with a better outcome may occur.
# Conclusion
This prospective, randomized, controlled trial confirms that lowdose exercise training at 4-7 days/week as add-on to PAH-targeted medication significantly improves peak VO 2 . The study indicates that exercise training improves cardiopulmonary parameters such as pulmonary vascular resistance and CI at rest and during exercise, leading to an increase in exercise capacity and QoL. The results of the study suggest that this therapy can improve RV function and other important prognostic parameters. The precise mechanism through which exercise-based interventions benefit PH patients still remains unclear. Further, larger multicentre trials using CI during exercise and time to clinical worsening as primary endpoints are necessary to confirm these results and to assess the impact of this therapy on patient outcomes.
[fig] Figure 1: Study design. The flowchart shows the number of allocated patients for each treatment group, the number of patients valid for analysis, and the number and reasons for exclusion, respectively (original data). [/fig]
[fig] 5: weeks in 74 of 87 (81.6%) patients: 34 patients of the training group and 40 patients of the control group agreed to invasive assessment of haemodynamics by RHC. Haemodynamics at rest and during exercise significantly improved in the training group, compared with the control group (Figure 3). The measurements at rest (Figure 3) revealed a significant improvement of CI (training +0.2 + 0.6 L/min/m 2 ; +9.3% vs. control 20.3 + 0.9 L/min/m 2 ; 26.5%, P , 0.001), CO (training +0.7 + 0.9 L/min/m 2 ; +12.5% vs. control 20.5 + 1.3 L/min/m 2 ; 25.3%, P , 0.001), mean pulmonary arterial pressure (mPAP) at rest (training 24 + 10 mmHg; 27.3% vs. control +4 + 8 mmHg; +16.1%, P ¼ 0.007), and pulmonary vascular .... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ..... .... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ..... .... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ..... .... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ..... .... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ..... .... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ..... [/fig]
[fig] Figure 2: Primary endpoint: change in peak oxygen uptake. Left graph: The abscissa shows peak VO 2 /kg at baseline, and the ordinate shows change of peak VO 2 /kg from baseline after 15 weeks for each patient. The solid points represent patients of the training group, and the circles represent patients of the control group. This graph shows equal distribution of baseline peak VO 2 in both groups as well as changes after 15 weeks of exercise training. Change in peak VO 2 did not correlate with baseline peak VO 2 . Right graph: the boxplots at the right side show median change of VO 2 max in the control vs. training group after 15 weeks. A significant increase can be shown in the training group (P , 0.001), whereas the control group shows a small reduction in peak VO 2 after 15 weeks. Multiple imputations showed the same P-value.Training improves cardiorespiratory function in PAH and CTEPH . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........ .... .. ...... ...... ...... ...... ...... ...... ...... ...... .. ... .... ... ... .... ... ... ... .... ... ... .... ... ... . .... .... ..... ..... ..... .... ..... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... .... ..... ..... ..... .... ..... ..... [/fig]
[fig] Figure 3: Secondary endpoints: haemodynamic function. Results from RHC for mean pulmonary arterial pressure (A), pulmonary vascular resistance (B), cardiac output (C) and cardiac index (D) at rest. The graphs depict the change of each parameter in per cent from baseline to 15 weeks after exercise training/no training. The mean changes of the training group, compared with baseline and control, as absolute values are given at the top of each graph with corresponding P-values of the ANCOVA for original data and multiple imputations. The bars are representing two times standard error. [/fig]
[fig] Figure 6: QoL. Mean changes of QoL + 2 standard errors of the mean and sample sizes below each bar. The bars of the control group are [/fig]
[table] Table 1: Baseline characteristics [/table]
[table] Table 2: Change echocardiography, CPET, and haemodynamics [/table]
|
Inferior Vena Cava and Renal Vein Thrombosis Associated with Thymic Carcinoma
Thymic tumors are rare mediastinal tumors that can present with a wide variety of symptoms. They can cause distant manifestations and are frequently associated with paraneoplastic syndromes. In our case, we describe the evolution of a 68-year-old male whose first manifestation was thrombosis of the inferior vena cava and renal veins. Thrombosis of large abdominal veins is rare, especially without being associated with any other comorbidity or risk factors.
# Introduction
Thymic tumors are rare mediastinal tumors that are considered orphan diseases due to their low prevalence. In the United States, they have an overall incidence of 0.13 cases per 100,000 person-year [bib_ref] Epidemiology of thymoma and associated malignancies, Engels [/bib_ref]. The most recent histologic classification divides thymic tumors into thymomas, thymic carcinomas (TC), and neuroendocrine thymic tumors (NETT) [bib_ref] Management of thymic tumors: a European perspective, Ruffini [/bib_ref]. Thymic tumors show a variable and unpredictable evolution, ranging from a benign form that can slowly grow for years before being discovered to a rapidly evolving metastatic tumor. They can often be asymptomatic, or the patient can present with local symptoms: chest pain, cough, and shortness of breath. When the tumor extends beyond its capsule and invades the surrounding tissue, then more severe symptoms can occur, such as superior vena cava (SVC) syndrome, hemidiaphragm paralysis (phrenic nerve involvement), and hoarseness (recurrent laryngeal nerve infiltration).
Thymic tumors are often associated with paraneoplastic syndromes, with myasthenia gravis being the most frequent [bib_ref] The spectrum of diseases associated with thymoma: coincidence or syndrome?, Souadjian [/bib_ref] [bib_ref] Thymoma and paraneoplastic myasthenia gravis, Marx [/bib_ref]. The list is long; however, one of the rarer manifestations can include a hypercoagulability state that can lead to thrombosis of the adjacent veins or even distal veins [bib_ref] Neoplastic and paraneoplastic vasculitis, vasculopathy, and hypercoagulability, Park [/bib_ref] [bib_ref] The link between cancer and venous thromboembolism: a review, Kessler [/bib_ref]. Abdominal vein thrombosis is a rare but potentially life-threatening form of venous thrombosis. Both inherited and acquired thrombophilia factors are frequently identified in cases of large vein thrombosis [bib_ref] Prothrombotic disorders in abdominal vein thrombosis, Leebeek [/bib_ref]. Advancements in molecular and genetic mechanisms have increased the spectrum of markers associated with a high risk of thrombophilia: mutation of FV Leiden 506R/Q, mutation of prothrombin (FII) 20210G/A, mutation of plasminogen activator inhibitor (PAI-1) 4G/5G, and others [bib_ref] Molecular pathophysiology of thrombotic states and their impact to laboratory diagnostics, Slavik [/bib_ref].
When a patient is diagnosed with a mediastinal mass, the differential diagnosis must include thymoma, lymphoma, thymic carcinoma, thymolipoma, germ cell tumors, and lung metastases but also thymic cysts, lymphangiomas, and aortic aneurysms [9, 10].
## Case presentation
We present the case of a 68-year-old male who is admitted to the regional hospital reporting pain in the upper abdomen.
The patient had no known history of any illness or any surgical intervention and did not take any medication at home. He was an ex-smoker (20 packs/year) but had quit smoking almost 25 years ago. His complaints started approximately a week before and progressively worsened until he decided to go to the hospital. The initial bloodwork showed slightly enhanced amylase = 174 U/L ( = 25-100 U/L) and lipase = 97 U/L ( = 0-60 U/L). EKG and chest radiography showed no modifications; however, the abdominal ultrasound revealed a right renal mass with modifications of the renal parenchyma and also modification of the pancreas ecostructure. The patient is transferred to our clinic with suspicion of right renal mass and pancreatitis.
The physical examination did not reveal much. The patient was in a relatively good state, with his only complaints being epigastric pain. His skin and mucosae were relatively pale, and he had telangiectasias on his face and chest. His respiration was normal, with 18 resp/min, and there were no audible rales. His blood pressure was 150/80 mmHg with 84 beats/min. His abdomen was slender and he reported slight pain on examination of the epigastrium.
The patient's bloodwork showed signs of anemia and thrombopenia which, in the 4th day after admission, reached the lowest level: 82.000/ul thrombocytes ( = 150-400.000/ul) and 11.1 g/dl of hemoglobin ( = 12-16 g/dl). He also had hypoalbuminemia, 3.1 g/dl ( = 3.5-5.1 g/dl), and proteinuria of 2.28 g/24 h.
The CT scan of the abdomen and pelvis proved to be an indispensable tool for the diagnosis. It showed thrombosis of the left and right renal veins and of the portal vein [fig_ref] Figure 1: CT scan cross section [/fig_ref] with enlarged kidneys. Also multiple perigastric, celiomesenteric, and hepatic adenopathies were described. He was immediately started on i.v. heparin which afterwards was replaced with oral anticoagulants.
The abdominal CT excluded the suspicion of a renal tumor. Having ruled out the most likely suspect that can cause renal vein thrombosis, we started to look for more unusual causes. All the tumor markers came back negative. But the extended coagulation testes revealed elevated Ddimers (980 g), present PDF, and a protein S deficiency (45%). However, the protein S deficiency alone is not enough to cause the massive vein thrombosis. We also tested for a genetic hypercoagulability state, which included tests for factors II, V, MTHFR, and PAI genes; all tests came back negative.
A thoracic CT scan was performed which revealed an anterior mediastinal mass in the upper quadrant, of approx 49/29 mm dimensions, with irregular shape and heterogenous iodophilia; it had small areas of necrosis; it was situated between the ascendant aorta and the trunk of the pulmonary artery and had lateral relation with the pulmonary parenchyma. Multiple mediastinal adenopathies could be seen [fig_ref] Figure 2: CT scan cross section [/fig_ref].
The patient was scheduled for surgical intervention. Sternotomy was performed followed by tumorectomy of the anterior mediastinal mass and thymomectomy of the remaining thymus tissue. The thymic tumor mass was 7/5/3 cm in size and was encapsulated being well differentiated from the surrounding tissue. Histopathology showed an undifferentiated carcinoma probably of thymic origin (optical microscopy showed poorly differentiated carcinoma cells with frequent atypical mitosis in the thymic mass which was covered by a hyaline capsule; the immunohistochemistry showed a CD5 positivity along with negative TTF1, negative PSA, negative Ck7, negative CROMO, negative SYNAPTO, and positive CK34BetaE12, P63, and PSMA). The operation was uneventful and the patient remained under observation for almost two weeks.
When he returned a month later for a follow-up, he was symptom-free and was feeling like a new man. The dyspnea and abdominal pain were gone and he was no longer feeling fatigued. An abdominal ultrasound revealed partial regression of the inferior vena cava thrombosis. The patient remained on oral anticoagulants and was followed up every 3 months. After one year, a control CT scan showed complete regression of the inferior vena cava and renal vein thrombosis [fig_ref] Figure 3: CT scan cross section [/fig_ref]. He was symptom-free and there was no need to continue taking acenocumarol.
# Discussion
Thymic tumors are rare mediastinal tumors and, most often, they are discovered by using imaging techniques (CT) because of nonspecific symptoms. However, in our case, thrombosis of the inferior vena cava and of both renal veins as the first manifestation of such tumors has never been described in the literature. The nonspecific symptoms of the patient (asthenia, dry cough, and pain in the upper abdomen) were what determined us to closely investigate the patient and perform an echography and, afterwards, a CT scan. But to our surprise, the cause of his symptomatology was neither abdominal nor pulmonary but mediastinal. The thymic carcinoma manifested itself as a paraneoplastic syndrome which induced a hypercoagulability state that led to the thrombosis of the inferior vena cava and renal veins. The modifications in the patient's blood can be attributed to the thrombosis which affected the connected organs (causing proteinuria, increased amylase), except for the anemia and the thrombopenia which might have been another paraneoplastic manifestation. The patient's hereditary thrombophilia (the protein S deficiency) also played an important part and was probably the reason why the first paraneoplastic manifestation presented as a thrombosis of major veins.
The initial therapeutic attitude after the diagnosis of inferior vena cava and renal vein thrombosis was the correct one. Firstly, heparin anticoagulation was initiated and was maintained until oral anticoagulants could be administered and had taken effect. This prevented the thrombosis from advancing and even managed to slightly regress in its spread. The thoracic surgeon performed ideal tumorectomy and according to the Masaoka-Koga staging system (1994), the thymic tumor was of stage I (grossly and microscopically completely encapsulated tumor) [bib_ref] Followup study of thymomas with special reference to their clinical stages, Masaoka [/bib_ref]. For this reason, no adjuvant chemotherapy or radiotherapy was necessary [bib_ref] Management of stage I and II thymoma, Detterbeck [/bib_ref] [bib_ref] The role of radiotherapy in the management of thymic tumors, Girard [/bib_ref] [bib_ref] Chemotherapy for thymic carcinoma and advanced thymoma in adults, Wei [/bib_ref].
One month after the excision of the tumor, the thrombosis of the inferior vena cava and of the renal veins had decreased (on Doppler echography). The symptomatology had completely disappeared and the patient remained on oral anticoagulants. He was followed up every 3 months and was scheduled for a CT scan after one year. When the followup CT scan was performed it showed complete regression of the inferior vena cava and renal vein thrombosis. Being symptom-free and with no sign of thrombosis, the patient was taken of oral anticoagulants.
# Conclusion
Thrombosis of the inferior vena cava and renal veins is a rare manifestation that can be caused by a handful of diseases. The list is short, with the most frequent being renal cancer. In our case, it was a manifestation of a paraneoplastic syndrome caused by thymic carcinoma. Although rare, thymic carcinomas can cause distant manifestations and are discovered when nonspecific symptoms start to emerge. In our case, the inferior vena cava and renal vein thrombosis without any local or apparent cause was what determined us to search further and investigate distant sites and rarer causes for this manifestation.
[fig] Figure 1: CT scan cross section: thrombosis of the left and right renal veins and of the portal vein. [/fig]
[fig] Figure 2: CT scan cross section: anterior mediastinal mass, upper quadrant, with irregular shape, heterogenous iodophilia; it has small areas of necrosis; axial dimension approx 49/29 mm and 50 mm craniocaudal; it is situated between the ascendant aorta and the trunk of the pulmonary artery and has lateral relation with the pulmonary parenchyma. [/fig]
[fig] Figure 3: CT scan cross section: complete repermeabilisation of the inferior vena cava and of the renal veins. [/fig]
|
Study of Rigid Cross-Linked PVC Foams with Heat Resistance
Three heat resistant cross-linked PVC foam plastics were prepared and their performances were compared with universal cross-linked PVC structural foam. The results show that these three heat resistant foams have higher glass transition temperatures (close to 100 °C) than universal structural foam (83.2 °C). Compared with the universal structural foam, the three heat resistant foams show much higher decomposition temperature and better chemical stability due to the crosslinking of PVC macromolecular chains. The heat distortion temperature (HDT) values of the three heat resistant foam plastics are just a little higher than that of universal structural foam. The three heat resistant foam plastics have good dimensional stability at 140 °C, and when used as core material can closely adhere to the face plates in medium temperature curing processes. Compared with universal structural foam, the three heat resistant foam plastics have slightly better mechanical properties.
# Introduction
Rigid cross-linked PVC foam is an ideal core material for sandwich structure composites because of its excellent features, such as outstanding stiffness and strength to weight ratios, self-extinguishing nature, good chemical resistance, sound and thermal insulation properties and low cost. Therefore, it is widely used in wind energy, marine, road and rail, aerospace, recreation and industrial applications [bib_ref] The preparation and properties of rigid cross-linked PVC foam, Shi [/bib_ref].
The vacuum-assisted resin infusion (VARI) process is a very popular advanced liquid composite molding process [bib_ref] Infusion design methodology for thick-section, low-permeability preforms using inter-laminar flow media, Heider [/bib_ref] , which has the outstanding advantages of low cost, being especially suitable for OPEN ACCESS large size production, high performance and low porosity of molded articles and environmental friendliness [bib_ref] Resin flow behavior simulation of grooved foam sandwich composites with the vacuum..., Zhao [/bib_ref]. So far, the VARI process has been widely used for forming ships, automobiles, aircraft, wind turbine blades and other structural parts [bib_ref] Optimization of the VARTM process for enhancement of the degree of devolatilization..., Grujicic [/bib_ref].
Currently the term commercial cross-linked PVC foams generally refers to rigid polymeric foams based on polyvinyl chloride (PVC), which is modified by an interpenetrating polymer network with aromatic amides (in this paper, called universal cross-linked PVC structural foam). The PVC molecular structure in this kind of foam is linear and its glass transition temperature (T g ) is about 80 °C [bib_ref] The structure and property relationships of commercial foamed plastics, Lin [/bib_ref]. This universal cross-linked PVC structural foam is fully suitable for room temperature VARI processes. In current industry practice, a medium temperature curing process (70~90 °C) is commonly used in order to improve the fluidity of the resin and the production efficiency. Heat released in the curing process under medium temperature conditions results in system temperatures of up to 120 °C, and even 140 °C. Universal cross-linked PVC structural foam core tends to shrink and release gas during the cure process in medium temperature processes, which means the core material can't adhere to glass fiber or other faceplates. Thus research on cross-linked PVC foam plastics with higher heat resistance has significant engineering application value.
There are several methods to improve the heat resistance of a PVC resin, such as copolymerization [bib_ref] Improvement in the heat resistance of poly(vinyl chloride) profile with styrenic polymers, Zhang [/bib_ref] , crosslinking [bib_ref] Crosslinking of plasticized PVC used in coated fabrics, Zadhoush [/bib_ref] , halogenation [bib_ref] Structure and properties of chlorinated polyvinyl chloride graft copolymer with higher property, Xu [/bib_ref] and blending modifications [bib_ref] Study of structural, morphological and mechanical properties of PMMA, PVC and their..., Kelkar [/bib_ref]. Crosslinking occupies an important position in numerous modification methods, and plays an active role in overcoming the defects of PVC, such as low softening point and poor dimensional stability at elevated temperatures [bib_ref] Chemical modification and characterization of poly(vinyl chloride) by crosslinking of multifunctional amines, Singh [/bib_ref]. Chemical crosslinking methods of PVC mainly include peroxide crosslinking [bib_ref] Peroxide crosslinking of rigid poly(vinyl chloride), Gunewardena [/bib_ref] [bib_ref] Peroxide crosslinking of rigid poly(vinyl chloride), Thomas [/bib_ref] [bib_ref] Peroxide crosslinking of plasticized poly(vinyl chloride), Saethre [/bib_ref] , silane crosslinking [bib_ref] Silane crosslinking of plasticized poly(vinyl chloride), Fiaz [/bib_ref] [bib_ref] Silane cross-linking of plasticized poly(vinyl chloride), Hearn [/bib_ref] and triazine compound crosslinking [bib_ref] Peroxide crosslinking of PVC foam formulations, Yanez-Flores [/bib_ref] , etc. Most studies are focused on the crosslinking of flexible PVC foam, and seldom does research pay attention to rigid PVC foam crosslinking because of the difficulties involved. In this paper, three heat resistant cross-linked PVC foam plastics were successfully prepared and compared with universal cross-linked PVC structural foam. Two kinds of foams are prepared by the following method: trimethylolpropane trimethacrylate (TMPTMA) or triallyl isocyanurate (TAIC) serving as crosslinking agent are added into a standard universal cross-linked PVC structural foam formula. The macromolecular chains of PVC are cross-linked by crosslinking agent under the action of initiator, and thus a foam plastic with an interpenetrating polymer network structure is formed by winding with the cross-linked network generated by the action of isocyanate, anhydride and water (marked as heat resistant cross-linked PVC foam I and heat resistant cross-linked PVC foam II, respectively). The third cross-linked PVC foam plastic is prepared by the following method: copolymerization of maleic anhydride (MAH) and acrylonitrile (AN) is followed by grafting with PVC macromolecular chains, and then reacted with isocyanate and water (marked as heat resistant cross-linked PVC foam III). The three foam plastics obtained have good heat resistance, and are very compatible with higher processing or service temperatures, so they should have extensive applications and a bright development future.
The preparation of universal cross-linked PVC structural foam has been reported in previous research [bib_ref] The preparation and properties of rigid cross-linked PVC foam, Shi [/bib_ref] and other literatures [bib_ref] Gradient foam core materials for sandwich structures: Preparation and characterisation, Danielsson [/bib_ref]. The reaction mechanism can be described as follows: a paste-like product was gelled in a molding process, while the gas generated by decomposition of the chemical foaming agent was dispersed in the gelation molded block, and a semi-foamed molded product was obtained after cooling. Then, in the boiled forming process, a polyurea/polyamide/polyimide crosslinking network was formed by the reaction of isocyanate, anhydride and water. This network embraced the PVC chains giving an entangled structure and a semi-interpenetrating polymer network was obtained. The decomposition mechanism of foaming agents have been reported [bib_ref] The preparation and properties of rigid cross-linked PVC foam, Shi [/bib_ref] [bib_ref] Thermal treatment and degradation of cross-linked ethylene vinyl acetate-polyethylene-azodicarbonamide-ZnO foams. Complete kinetic..., Reyes-Labarta [/bib_ref] [bib_ref] Kinetic study of the decompositions involved in the thermal degradation of commercial..., Reyes-Labarta [/bib_ref] , and the schematic diagram of the boiled foaming reaction mechanism is depicted in Scheme 1. Peroxide and crosslinking agent are introduced into heat resistant cross-linked PVC foam I and II. Except for the same reaction of universal cross-linked PVC structural foam in the boiled foaming process, peroxide crosslinking of PVC also exists in the molded process. There is a lot of research focused on the peroxide crosslinking of PVC to improve its heat resistance [bib_ref] Peroxide crosslinking of rigid poly(vinyl chloride), Gunewardena [/bib_ref] [bib_ref] Peroxide crosslinking of rigid poly(vinyl chloride), Thomas [/bib_ref] [bib_ref] Peroxide crosslinking of plasticized poly(vinyl chloride), Saethre [/bib_ref] [bib_ref] Peroxide crosslinking of unplasticized poly(vinyl chloride), Garcia-Quesada [/bib_ref] , and the crosslinking mechanism is well known, but it has not been used in the preparation of rigid low density cross-linked PVC foam plastics. Vinyl monomer and unsaturated anhydride are introduced in heat resistant cross-linked PVC foam III. Except for the similar reactions among of isocyanate, anhydride and water to universal cross-linked PVC structural foam, the graft copolymerization also exists. Scheme 2 shows the general scheme of the graft copolymerization mechanism. In addition, the nitrile group induced by the graft copolymer will be hydrolysed in the boiled foaming process, as shown in Scheme 3.
## Scheme 1.
The schematic diagram of the boiled foaming reaction mechanism of universal cross-linked PVC structural foam.
[formula] R 1 NCO OCN + H 2 O 2 H 2 N R 1 NH 2 + CO 2 2 R 2 C C O O O + H 2 O R 2 C C O O OH OH H 2 N R 1 NH 2 + R 1 NCO OCN H 2 N R 1 NH R 1 NCO NH C O H 2 N R 1 NH 2 + R 2 C C O O OH OH R 2 C C O O NH NH R 1 R 1 NH 2 NH 2 2 H 2 N R 1 NH 2 R 2 C C O O O + R 2 C C O O N R 1 NH 2 + H 2 O Scheme 2. [/formula]
The schematic diagram of the molded process reaction mechanism of heat resistant cross-linked PVC foam III.
[formula] CH 2 CH CN n H 2 O CH 2 CH CONH 2 CH 2 CH C CH 2 b CH C O O N H c CH 2 CH CN a [/formula]
# Results and discussion
## Gel content measurement
Gel content measurements were conducted to determine the degree of crosslinking of the samples. All foam plastics were controlled with the foam density of 70 kg/m 3 for comparison, and the gel content results are shown in [fig_ref] Table 1: Gel contents of molded products and foam plastics [/fig_ref]. As shown in [fig_ref] Table 1: Gel contents of molded products and foam plastics [/fig_ref] , the gel content of molded products of universal cross-linked PVC structural foam is 0, while that of foam plastics is 49.3%. This result implies that the crosslinking reaction has occurred in the boiled foaming process, but not in the molded process. The gel contents of molded products of heat resistant cross-linked PVC foam I and II are not 0, and that of foam plastics of heat resistant cross-linked PVC foam I and II are much higher than that of universal cross-linked PVC structural foam. This result implies that the degree of crosslinking is increased by the peroxide crosslinking of PVC in the molded process. The gel contents of the heat resistant cross-linked PVC foam III molded product is 0, and that of foam plastic of heat resistant cross-linked PVC foam III reaches as high as 96.4%. This result implies that a highly cross-linked heat resistant PVC foam III is realized by the graft copolymerization.
## Thermal mechanical analysis (tma)
TMA tests were carried out for the determination of T g , which is related to the foam density. Therefore, the foam densities were controlled the same for comparison. TMA curves of heat resistant cross-linked PVC foam I, II, III and universal cross-linked PVC structural foam with the foam density of 70 kg/m 3 are shown in . As shown in the figure, all the TMA curves begin to deform at 60 °C. T g is obtained from the intersection of the bitangent of the curves. According to the TMA results, T g values of universal cross-linked PVC structural foam and heat resistant cross-linked PVC foams I, II, III with a foam density of 70 kg/m 3 are 83.2 °C,101.4 °C, 99.7 °C and 103.8 °C, respectively. The results show that T g values of foam plastics are significantly increased and heat resistant performance improves after the crosslinking of the PVC chains. In other words, the T g values obtained by TMA as the upper limit of continuous operating temperature is enhanced due to the crosslinking of PVC chains. Ⅱ Ⅲ
Heat Flow / (W/g) Temperature /℃ Endothermic Ⅰ universal resistant cross-linked PVC foam I and II, the endothermic peak caused by PVC decomposition is decomposed into two peaks. This can be explained by that two forms of PVC exist in heat resistant cross-linked PVC foams I and II, one of which involves linear PVC penetration in the cross-linked network, while the other is cross-linked PVC. The peak temperature of the endothermic area caused by PVC decomposition in heat resistant cross-linked PVC foam III is 284 °C, which is related to the almost completely cross-linked structure. This matches well with the former gel content measurements and TMA tests. It is obvious in [fig_ref] Figure 3: demonstrates the TG/DTG analysis for heat resistant cross-linked PVC foams I, II,... [/fig_ref] and [fig_ref] Table 2: TGA values of foam plastics [/fig_ref] that all of the three kind heat resistant cross-linked PVC foams have higher T d 5 , T d 10 , T d , T d 50 and residual weight than the universal cross-linked PVC structural foam.
## Thermogravimetric analysis (tga) and derivative thermogravimetry (dtg) analysis
The decomposition temperature and weight loss temperature of heat resistant cross-linked PVC foam III improved significantly, while heat resistant cross-linked PVC foam I and II just improve slightly. This phenomenon indicates that the crosslinking of PVC chains is beneficial to heat resistance performance. The descending order of heat stability of the four kinds of foams is summarized as follows: heat resistant cross-linked PVC foam III, heat resistant cross-linked PVC foam II, heat resistant cross-linked PVC foam I, universal cross-linked PVC structural foam.
## Heat distortion temperature (hdt) analysis
The HDT of foam plastic increases with elevated foam density. Therefore, the density was kept at 70 kg/m 3 for all four foams for comparison. [fig_ref] Figure 4: Heat distortion temperature curves of foam plastics [/fig_ref] shows the corresponding HDT curves. As shown in [fig_ref] Figure 4: Heat distortion temperature curves of foam plastics [/fig_ref] , the thickness of universal cross-linked PVC structural foam increased from room temperature and then decreased after 95 °C. However, the thicknesses of heat resistant cross-linked PVC foams I, II, III are all increased until the temperature reaches 120 °C, which is about 25 °C higher than in universal cross-linked PVC structural foam. This phenomenon is related to the higher T g values of these three kinds heat resistant cross-linked PVC foams than that of universal cross-linked PVC structural foam. The heat expansion degree of gas in cells is so large that the thickness increases significantly as the temperature increases. Foam plastics reach a high elastic state when the temperature exceeds T g , and foam plastics are compressed caused by gas escape through the cell walls. When the temperature is high enough (exceeding 200 °C), the matrix resin of cell walls and cell struts undergo further crosslinking curing, which means the gas cannot escape from the cells and this causes a secondary expansion.
According to the DIN53424 standard, for the rigid, closed-cell foam plastics the temperature when the compression deformation reaches to 2 mm is defined as HDT. However, in this work, the thickness decrease under compressive load cannot reach 2 mm, so the temperature when the thickness of foam decreases to the lowest point is considered as HDT. As shown in [fig_ref] Figure 4: Heat distortion temperature curves of foam plastics [/fig_ref]
# Dimension stability analysis
The test for dimension stability of foam plastics with foam density of 70 kg/m 3 was implemented according to ISO 2796. At 160 °C, brittle rupture will take place when foam plastics expand to a certain extent. The heating time (at 160 °C) for heat resistant cross-linked PVC foams I, II, III and universal cross-linked PVC structural foam to begin rupturing are 12 h, 14 h, 16 h and17 h, respectively. Therefore, the data of weight and volume change rate at the point of 1h before the rupture is selected. [fig_ref] Table 3: Dimension stability of foam plastics [/fig_ref] lists the dimension stability values of the different foam plastics. As shown in [fig_ref] Table 3: Dimension stability of foam plastics [/fig_ref] , the dimensional stability rule of heat resistant cross-linked PVC foam under high temperature conditions is consistent with that of universal cross-linked PVC structural foam. Volume shrinkage occurs in all foams at temperatures below 120 °C, and then the shrinkage degree decreased with elevating temperature. Brittle rupture will take place in the foam plastics after excessive expansion at 160 °C. When the temperature exceeds the T g of the foam plastics, the weight loss rate and volume shrinkage rate of universal cross-linked PVC structural foam increase rapidly, while they are small for the three kinds of heat resistant cross-linked PVC foam under the same temperature conditions. It was also found that the weight and volume change rate of the three heat resistant cross-linked PVC foam increase slightly more than that of universal cross-linked PVC structural foam with elevating temperature. This phenomenon may be because that those three kinds heat resistant cross-linked PVC foam have higher T g values, and the chains in the cell wall and cell struts are difficult to move below 120 °C, so the foam shrinkage is avoided. Further curing may take place in matrix resin of cell walls and cell struts when the temperature is higher than 140 °C, thus the volume change rate of the three kinds of heat resistant cross-linked PVC foam is small when the heating temperature is below 140 °C. Furthermore, the heating time before brittle rupture of these three kinds of elevated temperature PVC structural foam is longer than that of universal cross-linked PVC structural foam. This is because the degradation temperature of the PVC resin in the heat resistant cross-linked PVC foams is elevated. As shown in [fig_ref] Table 4: demonstrates the mechanical properties of cross-linked PVC foam plastics with a density... [/fig_ref] , the mechanical properties of heat resistant cross-linked PVC foam I and II are slightly higher than universal cross-linked PVC structural foam. This may be because all these foam plastics have interpenetrating polymer network structures. Under the condition of similar cell structure, the mechanical properties are decided by the performance of the matrix resin. Although the crosslinking of PVC chains is conducive to improving the strength of the matrix resin, the strength of the polyurea/polyamide/polyimide network is not changed. In addition, only slight improvement is observed in heat resistant cross-linked PVC foams I and II. Compressive strength and modulus, tensile strength and modulus and shear strength of heat resistant cross-linked PVC foam III are slightly higher than that of universal cross-linked PVC structural foam. However, the elongation at break in the tensile and shear strength testing is reduced. This may be because the PVC chains in heat resistant cross-linked PVC foam III are cross-linked by a copolymer of MAH and AN, and the high degree of crosslinking leads to less flexibility and less extension. [fig_ref] Table 5: Formulas of cross-linked PVC foam plastics [/fig_ref] gives the formulas of three heat resistant cross-linked PVC foams and universal cross-linked PVC structural foam. Foam plastics are prepared as follows: firstly, all materials are blended until a perfectly homogeneous paste-like product is formed. Secondly, the paste-like product is molded for 20 min under the conditions of 170 °C and 15 MPa. Thirdly, the molded product is exposed to a steam of hot water (97 °C) to expand until the desired density is obtained. Finally, foam plastics are treated in warm water (50 °C) to remove any residual active ingredients. [fig_ref] Table 1: Gel contents of molded products and foam plastics [/fig_ref] 3.3. Characterization
# Mechanical properties analysis
## Experiment preparation
## Gel content measurement
The degree of crosslinking can be estimated through the gel fraction. The gel contents of molded products and foam plastics were determined gravimetrically by Soxhlet extraction for 24 h using tetrahydrofuran (THF) as the solvent. Approximately 0.5 g of the molded product or foam plastic (m 0 ) was cut into small pieces. After the extraction cycle, the sample was dried to constant weight (m 1 ). Taking into account the initial insoluble fraction (f) of the sample, the gel content (GC) can be calculated according to expression (1):
[formula] 1 0 % (1 ) 100 (1 ) m GC m f (1) [/formula]
# Heat resistance analysis
The glass transition temperature (T g ) of foam plastic was obtained using an Mettler-Toledo TMA 840 thermomechanical analyzer. The temperature was risen from 30 °C to 140 °C at the speed of 5 °C/min, and the sample size for TMA analysis was 10 mm × 8 mm × 6 mm. Heat transformation of foam plastic was recorded by a TA Q2910 differential scan calorimeter. The temperature was increasedfrom 25 °C to 300 °C at a rate of 10 °C/min. Weight change and thermal decomposition of foam plastic were analyzed by a NETZSCH 209 F1 thermogravimetric analyzer from 50 °C to 650 °C at the rate of 10 °C/min. The atmosphere used was nitrogen with a flow rate of 30 mL/min. HDT of foam plastic was measured by a Martin heating test box 110A which was produced by the instrument repair center of Northwestern Polytechnical University. Dimensional stability of foam plastic was measured in a 101-2A drying oven under forced convection. The dimensional stability test is implemented according to ISO 2796.
# Mechanical property analysis
Mechanical properties of foam plastics were measured by an MTS CMT7204 universal testing machine. Flatwise compressive properties, tensile properties, shear properties of foam plastics were tested according to standards ASTM D1621, D638, C273, respectively. At least five specimens were tested for each sample.
# Conclusions
In this work, three heat resistant foam plastics are prepared and their performances are compared with universal cross-linked PVC structural foam. The T g values of heat resistant cross-linked PVC foams I, II, III are 101.4 °C, 99.7 °C and 103.8 °C, respectively, which are higher than that of the universal structural foam. The crosslinking of PVC macromolecular chains makes heat resistant cross-linked PVC foams I, II, III have higher decomposition temperatures and better heat stability than universal structural foam. HDT values of heat resistant cross-linked PVC foams I, II, III are just a little higher than that of universal cross-linked PVC structural foam. The three heat resistant foam plastics all have good dimensional stability and small volume change rates. The volume change rates of the three heat resistant foam plastics are just less than half of that of the universal structural foam at 140 °C. Compared with universal cross-linked PVC structural foam, these three heat resistant cross-linked PVC foam plastics have slightly better mechanical properties.
[fig] Figure 1 . 3: TMA Differential Scanning Calorimetry (DSC) AnalysisFigure 2 shows the DSC analysis results for heat resistant cross-linked PVC foams I, II, III and universal cross-linked PVC structural foam. [/fig]
[fig] Figure 2: DSC curves of foam plastics.As shown inFigure 2, the peak temperature of the endothermic area caused by PVC decomposition in the universal cross-linked PVC structural foam is 259 °C. According to the DSC curves of heat [/fig]
[fig] Figure 3: demonstrates the TG/DTG analysis for heat resistant cross-linked PVC foams I, II, III and universal cross-linked PVC structural foam. The parameters of 5% weight loss temperature (residual weight are reported inTable 2. [/fig]
[fig] Figure 4: Heat distortion temperature curves of foam plastics. [/fig]
[table] Table 1: Gel contents of molded products and foam plastics. [/table]
[table] Table 2: TGA values of foam plastics. [/table]
[table] Table 3: Dimension stability of foam plastics. [/table]
[table] Table 4: demonstrates the mechanical properties of cross-linked PVC foam plastics with a density of 70 kg/m 3 .. Mechanical properties of foam plastics. [/table]
[table] Table 5: Formulas of cross-linked PVC foam plastics. [/table]
|
Pregnancy complications among nulliparous and multiparous women with advanced maternal age: a community-based prospective cohort study in China
Background: This study aimed to evaluate the incidence rates and risks of pregnancy complications among nulliparous and multiparous women with advanced maternal age (AMA, ≥35 years) in China. Methods: We performed a community-based prospective cohort study of 10,171 pregnant women in selected two sub-districts and 11 towns of Liuyang from 2013 to 2015. All subjects were followed up from the first prenatal care (at ≤12 weeks) to delivery, and risks of pregnancy complications were compared by parity and maternal age groups.Results: Among nulliparas, women with AMA showed significantly increased risks for gestational hypertension (OR 8.44, 95%CI 1.68-2.88), preeclampsia/eclampsia (OR 9.92, 95%CI 4.87-18.78), premature rupture of membrane (OR 6.84, 95%CI 2.00-17.69), as compared to women in the 20-29-year age group. Among multiparas with AMA, increased risks were found for gestational diabetes mellitus (OR 3.29, 95%CI 1.76-5.94), anemia (OR 1.85, 95%CI 1.25-2.69), polyhydramnios (OR 3.29, 95%CI 1.56-6.64), premature rupture of membrane (OR 5.14, 95%CI 2.12-12.29), and preterm labor (OR 1.89, 95CI 1.42-2.50). Conclusions: Women with AMA were associated with increased risks of pregnancy complications, and complications with increased risks differed in nulliparas and multiparas. Women with AMA should be identified as a high-risk group in clinical practice.
# Background
Since the 1970s, the Chinese government has implemented the one-child policy to control the rapid growth of the population for around 40 years. With emerging problem of the aging population, the population policy has gradually switched to the two-child policy which encourages the birth of a second child since 2011 and has been fully implemented since 2016. With the impact of this new population policy, China is expected to see a sharp increase in fertility rate, and the proportion of women with advanced maternal age (AMA) is estimated to increase significantly. Women with AMA may have increased risk for complications during pregnancy due to decreased ovarian and uterine function, as well as other concomitant diseases, which may result in threats to maternal and child health. Thus, the prevention and control of complications in women with AMA during pregnancy have become a major challenge in clinical practice.
Previous studies have reported that women with AMA had higher incidences of pregnancy complications, such as hypertensive disorders of pregnancy, gestational hypertension, placental disorders, preterm labor, maternal near miss, and maternal death. Moreover, research also suggested that the elevated risks of pregnancy complications in AMA may differ by parity, and inconsistent results were reported regarding certain complications. For instance, some studies found that the risk of gestational diabetes and hypertensive disorders of pregnancy were increased in both nulliparous and multiparous women with AMA, while in some studies the increased risks for the two complications were not observed in both nulliparas and multiparas. It is therefore important to examine the association between AMA and pregnancy complications by parity within different populations and consider various confounding factors, such as disease history and lifestyle factors.
To our knowledge, only a few studies analyzed pregnancy complications and AMA in China previously, and these studies had certain limitations. The majority of previous studies were case-control or cross-sectional studies, and most studies did not conduct stratified analysis by parity. Although there were two retrospective cohort studies that analyzed pregnancy complications and AMA in China, these two studies only recruited patients with pregnancy complications in hospitals, thus incidence rates of different pregnancy complications among different age groups were unknown. Also, previous studies mainly focused on the population lived in urban areas, while few studies focused on women lived in rural areas. These women usually have a low educational level and low socioeconomic status, who may need more support from the healthcare system on health education and prenatal care.
Our study aimed to evaluate the incidence rates and risks of pregnancy complications among women with different parity and maternal age groups through a community-based prospective cohort study. This study may highlight the need for prevention of pregnancy complications among women with AMA in rural areas.
# Methods
## Study population
Subjects for this study were recruited from multiple maternal and child healthcare centers in Liuyang city, Hunan Province, China from June 2013 to December 2015. Liuyang is a county-level city located in the central-south region of China. It has four sub-districts and 33 towns under its jurisdiction. In this study, we randomly selected two sub-districts and 11 towns in Liuyang city to represent the population from both urban and rural areas, and the recruitment was conducted in the maternal and child healthcare centers located in each selected sub-district/ town. Pregnant women were recruited by their gynecologists during their regular prenatal care visits with the following inclusion criteria: a) received the first prenatal care at the recruited healthcare center; b) had no more than 12 weeks of gestation. Subjects were excluded if they a) had pregnancy complications before recruitment in the first trimester of the index pregnancy because it's unclear whether the disease was started from this pregnancy or before this pregnancy; b) had an induced abortion during follow-up or did not have complete medical records from enrollment through delivery; c) had language communication barriers or were deaf; or d) refused to actively cooperate for this research.
Informed written consent was obtained from all participants. Ethical approval was obtained from the Ethics Committee of Xiangya School of Public Health, Central South University, and all research was performed in accordance with relevant guidelines/regulations.
## Follow-up and data collection
Baseline information was collected for all subjects by trained research gynecologists through a face-to-face interview.
The questionnaire included sociodemographic characteristics (i.e. age, education, occupation, family income, residential area), pre-pregnancy weight and height, and history of gravidity, parity, and abortion. Routine obstetric examinations and laboratory examinations were also conducted afterward. Subjects were followed up at 16 weeks, 24 weeks, 28 weeks, 32 weeks, 38 weeks of pregnancy (or before delivery), and the end of delivery, and routine obstetric examinations were conducted at each follow-up. During the follow-up period, medical records including the records of routine obstetric care, the diagnosis of pregnancy complications, and the time of diagnosis were collected. A supervisory team was assigned to check whether the follow-up was on time, verify the registered complications, and check and rearrange the missing diagnosis.
## Variable definitions
Subjects were categorized into four maternal age groups, including 20-25 years, 25-29 years, 30-34 years, and ≥ 35 years. Women with AMA were defined as age greater or equal to 35 years old. Subjects aged below 20 years old were excluded in stratified analysis, due to small sample size (n = 101). Residential area was categorized into urban and rural residence. Family per capita annual income was classified into four categories (in Chinese Yuan, CNY): ≤10,000, 10,001-20,000, 20,001-30,000, and ≥ 30,001. Education levels were classified into two categories: junior high school or lower and high school/ college. Occupations included farmer, housewife, factory worker, and others.
Pre-pregnancy body mass index (BMI) was calculated using the formula weight (kg)/height (m) 2 . BMI was categorized into three categories using Asian-specific cutoffs [19]: < 18.5 kg/m 2 , 18.5-23 kg/m 2 , 23-27.5 kg/m 2 , and ≥ 27.5 kg/m 2 . Gravidity and parity were categorized as none, once or more pregnancies/births. History of miscarriages was classified into three categories: none, once, twice and more times, and history of induced abortion and history of preterm labor was categorized as none and once or more times.
Gestational hypertension was defined as having high blood pressure after 20 weeks according to the International Society for the Study of Hypertension in Pregnancy guidelines. Chronic hypertension before 20 weeks of gestation was not an outcome of interest in this study. Gestational diabetes mellitus (GDM) was diagnosed by oral glucose tolerance test (OGTT) conducted at 24-28th week of gestation, with fasting plasma glucose > 5.6 mmol/l or with post-75 g glucose load glucose level of > 8.8 mmol/l in 60th min or > 7.8 mmol/l in 120th min. Anemia was defined as hemoglobin < 12.0 g/dL. Polyhydramnios and oligohydramnios were examined using ultrasound, and was defined as the 4-quadrant amniotic fluid index (AFI) > 24 cm or ≤ 5 cm, respectively. Threatened abortion was defined as a small amount of vaginal bleeding, followed by paroxysmal lower abdominal pain or low back pain before the 28th week of pregnancy. The pelvic examination was not open, the membrane was intact, no pregnancy was discharged, and the size of the uterus was consistent with the gestational age. Preterm labor was defined as birth occurs before the start of the 37th week of pregnancy. Low birth weight is defined as the birth weight of an infant of 2499 g or less.
# Statistical analysis
The demographic and pre-pregnancy characteristics of study participants were compared among four maternal age groups. Significant levels were compared using Chisquare tests, and all the tests were two-tailed. Incidence of pregnancy complications and mode of delivery were compared by parity and age groups, and we used subjects age 20 to 29 years as the reference group considering the low incidence of pregnancy complications in lower age groups. Logistic regression was used to calculate the odds ratio (OR) and 95% confidence interval (CI) for pregnancy complications. Univariate analyses were conducted on all socio-demographic and prepregnancy characteristics to select the variable for adjustment, and significant variables were included in the multivariable model. Cramer's V statistics were used to evaluate the correlation between variables included in the final model. We were not able to conduct logistic regression for placenta previa, infectious disease, and threatened abortion among nulliparas due to low incidence rate. In the multivariate analysis, income level and BMI were adjusted for gestational hypertension, preeclampsia/eclampsia, and threatened abortion; occupation was adjusted for GDM and premature rupture of membrane; income level was adjusted for anemia, polyhydramnios, oligohydramnios, placenta previa, and infectious disease; occupation and history of preterm labor was adjusted for preterm labor; education and BMI were adjusted for the mode of delivery and low birth weight. All analyses were conducted using R version 3.5.2.
# Results
## Socio-demographic and pre-pregnancy characteristics
A total of 12,170 pregnant women were admitted to the hospital from June 2013 to December 2015. Among them, 10,475 pregnant women were recruited in the study (response rate 86.1%), and 10,171 subjects with complete information were included in the final analysis (97.1%).
Among the 10,171 subjects, the average age was 26.7 ± 4.2 years. 3274 (32.2%) subjects were 20-24 years old, 4587 (45.1%) subjects were 25-29 years old, 1637 (16.1%) subjects were 30-34 years old, and 572 (5.6%) subjects were 35 years old and older . The majority of subjects (85.9%) lived in rural areas, and the median family annual income was 20,000 Chinese Yuan per capita (around 2880 USD). The majority of subjects did not complete high school education (83.7%), and most of them were housewives (54.3%), factory workers (23.5%), and farmers (13.9%). The pre-pregnancy BMI was normal for 61.4% of subjects, 18.0% of subjects were underweight, and 20.5% of subjects were overweight/obese. Overall, around half of the subjects had one or more births previously, and the proportions were much higher for women aged 30-35 years (79.0%) and women aged 35 years and above (88.6%) than younger age groups. Also, around onethird of women aged 35 years and above had two or more times of miscarriages, while the proportion ranged from 6.3 to 19.5% in younger age groups. 3.1% of subjects had a history of induced abortion, and the proportion ranged from 1.7 to 5.9% in different age groups.
Incidence rates of pregnancy complications by parity and maternal age groups
The comparison of incidence rates for different pregnancy complications stratified by parity and age groups is shown in. Overall, the incidence rate of each complication during pregnancy ranged from 0.9 to 4.2%, and the top three pregnancy complications observed included gestational hypertension (4.2%), preeclampsia or eclampsia (3.5%), and anemia (3.7%). The incidence of pregnancy complications showed an increasing pattern when stratified by age groups for gestational hypertension, preeclampsia, polyhydramnios, premature rupture of membrane among nulliparas, and this pattern was observed for GDM, anemia, polyhydramnios, oligohydramnios, placenta previa, premature rupture of membrane among multiparas. The incidence of preterm labor was 9.4%, which ranged from 9.3 to 13.8% and 8.1 to 14.5% among different age groups in nulliparas and multiparas, respectively. The incidence of low birth weight was 2.8%, and the incidence was higher in multiparas (5.2%) than nulliparas (0.7%). In terms of mode of delivery, around half of subjects aged 30 years old and above performed a cesarean delivery, while this proportion was around 30% in the 20-29-year age group. ORs for pregnancy complications in different maternal age groups by parity
Adjusted ORs for pregnancy complications calculated using logistic regression models are shown inWomen aged 20-24 years and 25-29 years were combined as one age group considering the low incidence of pregnancy complications in the two groups
# Discussion
In this study, we found that women with AMA had a significantly higher incidence of pregnancy complications than women with younger age, and the increased risks of pregnancy complications varied by parity. More specifically, nulliparous women with AMA showed increased risks for gestational hypertension, preeclampsia/ eclampsia, and premature rupture of membrane, while multiparous women with AMA showed increased risks for GDM, anemia, polyhydramnios, premature rupture of membrane, and preterm labor. Cesarean delivery was also associated with older maternal age, where increased risks were found for women aged 30-35 years and 35 years and above in both nulliparas and multiparas compared to women aged 20-29 years. We found that increased risks of gestational hypertension and preeclampsia/eclampsia for AMA were only observed in nulliparous women, not in multiparous women. Our finding is consistent with many studies that reported preeclampsia was more common in nulliparas, while inconsistent with some previous studies that found AMA was a risk factor for hypertensive disorders during pregnancy in both nulliparous and multiparous women. On the other hand, one study also suggested that AMA was not associated with preeclampsia irrespective of parity. Our findings may be explained by several factors. First, studies have suggested that hypertensive disease in nulliparous and multiparous pregnant women may involve different pathophysiology, especially in terms of immune maladaptation, though no conclusion has been made so far. Moreover, it's possible that the increased risk for the multiparous women with AMA was confounded by other risk factors, including history of hypertensive disorders in the first pregnancy, co-existence of other complications, and unhealthy lifestyle. Future studies may be needed to examine the potential confounding factors, as well as the pathophysiology of pregnancy complications in women with different parity, to explore reasons behind this phenomenon.
Consistent with previous studies, the risks of GDM, anemia, polyhydramnios were increased in women with AMA compared to women aged 20-29 years in the multiparous group. GDM is a common pregnancy complication for women with AMA, and most scholars believe that the increased incidence may be due to change in blood volume, vascular endothelial injury, insulin receptor and insulin affinity decreased with aging. Also, it has been reported that multiparity was associated with an increased risk of GDM, although the effects of increasing parity on insulin sensitivity or βcell function were not detected. Anemia during pregnancy is usually related to inadequate diets or not receiving prenatal iron and folate supplements.
Previous studies also reported that anemia was an independent risk factor for low birth weight and preterm delivery. Considering the prevalence of underweight and low education level in this area, it's important to provide health education and support to prevent and control anemia especially for women with AMA in healthcare practice.
A national survey reported that the overall cesarean delivery rate in China had increased from 28.8% in 2008 to 34.9% in 2014, with the rates increased linearly in rural and general urban areas but declined beyond baseline rates in super cities. In this study, the overall cesarean delivery rate was 35.5%, and AMA was found to be a risk factor for cesarean delivery, which is consistent with previous research. With the change in child policy, the health care system should be prepared to cope with the increasing challenge of women with AMA and history of cesarean delivery.
This study has some limitations. First, subjects in this study were recruited at about 12 weeks of gestation based on the regulation of pregnant women in the health management registration system, thus we might underestimate the incidence rates of complications during early pregnancy. Also, it is possible that some subjects did not participate in all of the screening tests suggested by their gynecologist, especially for those with low education and income level, which may result in a lower incidence rate. However, the results of our study reflected the incidence of pregnancy complications in routine clinical practice. Second, only a few subjects in extreme age groups were included in the study, thus we were unable to conduct stratified analysis for subjects aged younger than 20 years old and subjects aged 40 years and above. Third, the results may be impacted by unmeasured confounding factors as we were not able to adjust for pre-pregnancy lifestyle and behaviors such as diet, smoking, and physical activity, although we have conducted adjusted models for sociodemographic factors and pre-pregnancy BMI.
# Conclusions
To our knowledge, this is the first study to identify incidences of pregnancy complications in nulliparous and multiparous Chinese women with AMA using a community-based prospective cohort study. Our findings suggest that women with AMA should be regarded as high-risk groups for pregnancy complications in the practice of maternal health care and maternal care management. Also, different risks of pregnancy complications were observed for nulliparous and multiparous women with AMA. This study bears importance in early prevention of pregnancy complications in the practice of population-based pregnancy health care. |
Re-emergence of interferon-α in the treatment of chronic myeloid leukemia
Treatment for chronic myeloid leukemia (CML) has evolved from chemotherapy (busulfan, hydroxyurea) to interferon-a (IFNa), and finally to tyrosine kinase inhibitors such as imatinib. Although imatinib has profoundly improved outcomes for patients with CML, it has limitations. Most significantly, imatinib cannot eradicate CML primitive progenitors, which likely accounts for the high relapse rate when imatinib is discontinued. IFNa, unlike imatinib, preferentially targets CML stem cells. Early studies with IFNa in CML demonstrated its ability to induce cytogenetic remission. Moreover, a small percentage of patients treated with IFNa were able to sustain durable remissions after discontinuing therapy and were probably cured. The mechanisms by which IFNa exerts its antitumor activity in CML are not well understood; however, activation of leukemia-specific immunity may have a role. Some clinical studies have demonstrated that the combination of imatinib and IFNa is superior to either therapy alone, perhaps because of their different mechanisms of action. Nonetheless, the side effects of IFNa often impede its administration, especially in combination therapy. Here, we review the role of IFNa in CML treatment and the recent developments that have renewed interest in this once standard therapy for patients with CML.
## Introduction to chronic myeloid leukemia
Chronic myeloid leukemia (CML) is characterized by excessive myeloid proliferation and a recurrent cytogenetic abnormality known as the Philadelphia chromosome (Ph). The abnormality results from a balanced translocation between chromosomes 9 and 22, t(9;22)(q34; q11.2), which fuses the breakpoint cluster region (BCR) gene on chromosome [bib_ref] Interferon-alpha therapy for chronic myelogenous leukemia, Wetzler [/bib_ref] to the ABL gene on chromosome 9. [bib_ref] The minute chromosome (Phl) in chronic granulocytic leukemia, Nowell [/bib_ref] [bib_ref] Letter: a new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by..., Rowley [/bib_ref] The resultant BCR-ABL oncogene encodes a constitutively active fusion BCR-ABL p210 oncoprotein. The activity of BCR-ABL is central to the pathogenesis of CML because it alters the proliferation, natural death processes and migration of the neoplastic cells. [bib_ref] A cellular oncogene is translocated to the Philadelphia chromosome in chronic myelocytic..., De Klein [/bib_ref] [bib_ref] Acute leukaemia in bcr/abl transgenic mice, Heisterkamp [/bib_ref] [bib_ref] Induction of chronic myelogenous leukemia in mice by the P210bcr/abl gene of..., Daley [/bib_ref] As a consequence, the leukemic clone gradually replaces normal hematopoiesis. Residual normal hematopoiesis is present in the vast majority of patients with CML, but since it is suppressed, most of the blood cells are Ph þ .
## Overview of treatments for cml
Until the early 1980s, CML therapy was based on busulfan or hydroxyurea, which had a negligible effect on the natural course of the disease. Talpaz et al. [bib_ref] Leukocyte interferoninduced myeloid cytoreduction in chronic myelogenous leukemia, Talpaz [/bib_ref] [bib_ref] Clinical investigation of human alpha interferon in chronic myelogenous leukemia, Talpaz [/bib_ref] carried out the first pilot clinical trial of partially pure IFNa for the management of CML followed by a larger study. The pivotal finding was that IFNa induced cytogenetic responses, which were more durable and reproducible than those induced by chemotherapy. Although initially used in the partially pure form, recombinant forms-a2a (Hoffman La Roche, Basel, Switzerland) and a2b (Merck & Co. Inc. (formerly Schering Plough), Whitehouse Station, NJ, USA)became the dominant IFNs used in clinical studies at doses similar to those used with partially pure IFNa (that is, 2-5 MU/m 2 daily). Recombinant IFNa therapy achieved response rates similar to those observed with the purified human product. [bib_ref] Interferon-alpha produces sustained cytogenetic responses in chronic myelogenous leukemia. Philadelphia chromosome-positive patients, Talpaz [/bib_ref] [bib_ref] Chronic myelogenous leukemia: a concise update, Kantarjian [/bib_ref] [bib_ref] Prolonged subcutaneous administration of recombinant alpha 2b interferon in patients with previously..., Ozer [/bib_ref] [bib_ref] Interferon alpha-2b as therapy for patients with Ph'-positive chronic myelogenous leukemia, Alimena [/bib_ref] [bib_ref] The Austrian Biological Response Modifier (BRM) Study Group. Treatment of chronic myelogenous..., Thaler [/bib_ref] [bib_ref] Long-term treatment of chronic myelogenous leukemia with different interferons: results from three..., Niederle [/bib_ref] Results from over 1500 randomized patients demonstrated that although both IFNa and chemotherapy (hydroxyurea or busulfan) could induce hematological responses in CML, IFNa significantly improved patient survival, with a 5-year survival rate of 50-59% compared with 29-44% for patients receiving busulfan or hydroxyurea. [bib_ref] The German CML Study Group. Randomized comparison of interferon-alpha with busulfan and..., Hehlmann [/bib_ref] [bib_ref] The Kouseisho Leukemia Study Group. A randomized trial comparing interferonalpha with busulfan..., Ohnishi [/bib_ref] [bib_ref] on behalf of the UK Medical Research Council's Working Parties for Therapeutic..., Allan [/bib_ref] In a study of 1303 IFN-treated patients, median survival was 8.2 years for low-risk patients, 5.4 years for intermediate-risk patients and 3.5 years for high-risk patients. [bib_ref] Writing Committee for the Collaborative CML Prognostic Factors Project Group. A new..., Hasford [/bib_ref] The studies of single-agent IFNa are summarized in [fig_ref] Table 1: Single-agent trials of IFNa [/fig_ref]. In an effort to improve outcomes, IFNa was also combined with chemotherapeutic agents, such as cytarabine, hydroxyurea and busulfan, and even with intensive chemotherapy regimens [fig_ref] Table 2: Combination trials of IFNa [/fig_ref]. [bib_ref] Intensive chemotherapy induction followed by interferon-alpha maintenance in patients with Philadelphia chromosome-positive..., Kantarjian [/bib_ref] [bib_ref] Treatment of advanced stages of Philadelphia chromosome-positive chronic myelogenous leukemia with interferon-alpha..., Kantarjian [/bib_ref] [bib_ref] Interferon-alpha therapy for chronic myelogenous leukemia, Wetzler [/bib_ref] [bib_ref] Randomized comparison of interferon a and hydroxyurea with hydroxyurea monotherapy in chronic..., Hehlmann [/bib_ref] [bib_ref] Treatment of Philadelphia chromosome-positive early chronic phase chronic myelogenous leukemia with daily..., Kantarjian [/bib_ref] [bib_ref] Combined interferon alfa-2a and cytosine arabinoside as firstline treatment for chronic myeloid..., Arthur [/bib_ref] [bib_ref] Efficacy and toxicity of IFN-a2b combined with cytarabine in chronic myelogenous leukaemia, Lindauer [/bib_ref] [bib_ref] for the French Chronic Myeloid Leukemia Study Group. Interferon alfa-2b combined with..., Guilhot [/bib_ref] [bib_ref] for the Italian Cooperative Study Group on Myeloid Leukemia. A randomized study..., Baccarani [/bib_ref] With the exception of cytarabine, the combination therapies were not usually superior to IFNa alone.
To improve its pharmacokinetic characteristics, IFNa has been attached to polyethylene glycol, which protects it from proteolytic breakdown. The resulting pegylated IFNa (PegIFNa) has been approved for the treatment of chronic hepatitis B and C and melanoma. In addition, a phase I trial evaluated escalating doses of PegIFNa-2a ± cytarabine in patients with IFNa-resistant chronic phase CML (CML-CP). [bib_ref] Phase I evaluation of a 40-kDa branched-chain long-acting pegylated IFN-a-2a with and..., Talpaz [/bib_ref] Dose-limiting toxicity was not observed at the highest dose of 630 mg per week. The safety profile of PegIFNa was similar to that of unmodified IFNa. While phase I trials suggested that PegIFNa induced better response rates compared with the unmodified form, [bib_ref] Phase I evaluation of a 40-kDa branched-chain long-acting pegylated IFN-a-2a with and..., Talpaz [/bib_ref] [bib_ref] Phase 1 study of polyethylene glycol formulation of interferon a-2B (Schering 54031)..., Talpaz [/bib_ref] randomized trials showed mixed results. [bib_ref] of the Pegasys CML Study Group. Phase II, randomized, multicenter, comparative study..., Lipton [/bib_ref] [bib_ref] for the PEG-Intron CML Study Group. Pegylated recombinant interferon alpha-2b vs recombinant..., Michallet [/bib_ref] By 1990, the constitutive tyrosine kinase activity of BCR-ABL was linked to the pathogenesis of CML. This discovery spurred the development of the molecular-targeted therapy imatinib (Gleevec/ Glivec; Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA), which proved selective for killing cells expressing the BCR-ABL protein. [bib_ref] Effects of a selective inhibitor of the Abl tyrosine kinase on the..., Druker [/bib_ref] Clinical trials with imatinib followed and demonstrated impressive response rates in patients who had not responded to IFNa therapy. Of the patients in late CML-CP, 95% achieved a complete [bib_ref] The Kouseisho Leukemia Study Group. A randomized trial comparing interferonalpha with busulfan..., Ohnishi [/bib_ref] 3-9 MU rIFNa-2a Abbreviations: CHR, complete hematological remission; rIFNa, recombinant interferon-a.
Role of interferon in chronic myeloid leukemia M Talpaz et al hematological remission (CHR), 60% a major cytogenetic response (MCyR) and 41% a complete cytogenetic response (CCyR). [bib_ref] Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase..., Druker [/bib_ref] [bib_ref] for the International STI571 CML Study Group. Hematologic and cytogenetic responses to..., Kantarjian [/bib_ref] Initiated in June of 2000, the phase III International Randomized Study of Interferon and STI571 (IRIS) was the first large-scale trial to compare imatinib (400 mg daily) with IFNa plus low-dose cytarabine, the standard of care at that time. [bib_ref] Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic..., O'brien [/bib_ref] The results in 1106 patients with newly diagnosed CML-CP demonstrated that imatinib was better tolerated and induced higher CHR and CCyR rates that resulted in longer progression-free survival than IFNa. The IRIS study did not report survival differences between the treatments because 90% of patients in the IFNa arm eventually crossed over to imatinib. However, comparison with IFNa-treated historical controls indicated a significant survival advantage with imatinib. [bib_ref] Imatinib mesylate therapy improves survival in patients with newly diagnosed Philadelphia chromosome-positive..., Kantarjian [/bib_ref] [bib_ref] Survival benefit with imatinib mesylate versus interferon-alpha-based regimens in newly diagnosed chronic-phase..., Kantarjian [/bib_ref] [bib_ref] Survival advantage from imatinib compared with the combination interferon-a plus cytarabine in..., Roy [/bib_ref] The US Food and Drug Administration subsequently approved imatinib for the treatment of newly diagnosed patients with CML-CP. In 2010, the Food and Drug Administration approved the second-generation tyrosine kinase inhibitors (TKIs) dasatinib (100 mg daily) and nilotinib (300 mg twice daily) as frontline therapies for patients with CML-CP. Recent studies indicate that both dasatinib and nilotinib are superior to standard-dose imatinib with regard to CCyR, major molecular response (MMR), and prevention of progression to accelerated and blast phases. [bib_ref] Nilotinib versus imatinib for newly diagnosed chronic myeloid leukemia, Saglio [/bib_ref] [bib_ref] Dasatinib versus imatinib in newly diagnosed chronic-phase chronic myeloid leukemia, Kantarjian [/bib_ref] Current therapeutic guidelines also recommend the use of nilotinib or dasatinib in patients intolerant or resistant to imatinib therapy and state that IFNa 'should no longer be considered as initial therapy for CML', but could be considered in 'rare patients unable to tolerate imatinib, dasatinib or nilotinib'. [bib_ref] Chronic myeloid leukemia: an update of concepts and management recommendations of European..., Baccarani [/bib_ref] Another treatment option for this patient population is hematopoietic stem cell transplantation, which was introduced in the 1970s. Although associated with significant morbidity and mortality, this therapy cured a substantial percentage of the patients with CML who qualified for it. In fact, transplantation is still perceived as the only curative treatment for CML. In 2007, the German CML Study Group conducted a randomized clinical trial comparing primary allogeneic hematopoietic stem cell transplantation with best available drug therapy (IFNa-based, but many patients were switched to imatinib over the course of the study) for patients with early CML-CP. [bib_ref] Drug treatment is superior to allografting as first-line therapy in chronic myeloid..., Hehlmann [/bib_ref] For the first 8 years, the drug treatment arm demonstrated better survival curves than those of the transplant arm. Beyond 8 years, the survival curves became less distinct. These results suggested that drug therapy should serve as first-line treatment for patients with CML-CP.
## Limitations of tki treatment in cml and emerging roles for ifna
The phenomenal outcomes of the IRIS study notwithstanding, a significant number of patients will require second-line therapy as a result of intolerance to or failure of imatinib therapy. After 8 years, 45% of patients in the IRIS trial randomized to receive imatinib were no longer receiving imatinib because of toxicity (6%), suboptimal response/failure (16%) or other reasons (23%). [bib_ref] International randomized study of interferon vs STI571 (IRIS) 8-year follow up: sustained..., Deininger [/bib_ref] Furthermore, a small fraction of patients taking imatinib continue to progress to accelerated or blast phase every year. [bib_ref] International randomized study of interferon vs STI571 (IRIS) 8-year follow up: sustained..., Deininger [/bib_ref] Another significant limitation of imatinib is the inability of most CML patients to discontinue therapy and maintain their remission. [bib_ref] Discontinuation of imatinib therapy after achieving a molecular response, Cortes [/bib_ref] Rousselot et al. [bib_ref] Imatinib mesylate discontinuation in patients with chronic myelogenous leukemia in complete molecular..., Rousselot [/bib_ref] reported that even though disease could not be detected for a median of 32 months in 12 CML patients who received imatinib, once therapy was stopped the BCR-ABL transcript was detectable in six of the patients within 1-5 months. Furthermore, the six patients who did not immediately relapse had previously taken IFNa for 29-152 months. Other studies, including IRIS, have also demonstrated improved outcomes with imatinib in patients who received or responded to prior IFNa treatment. [bib_ref] High rates of durable response are achieved with imatinib after treatment with..., Guilhot [/bib_ref] [bib_ref] Imatinib mesylate therapy in chronic myeloid leukemia patients in stable complete cytogenic..., Alimena [/bib_ref] [bib_ref] Efficacy and safety of imatinib in patients with chronic myeloid leukemia and..., Branford [/bib_ref] [bib_ref] Discontinuation of imatinib in Japanese patients with chronic myeloid leukemia, Takahashi [/bib_ref] More recently, Mahon et al. [bib_ref] Discontinuation of imatinib in patients with chronic myeloid leukaemia who have maintained..., Mahon [/bib_ref] demonstrated that among patients who sustained a complete molecular response to imatinib for at least 2 years (a subset of patients that constitutes a minority of all treated patients), 40% did not relapse when therapy was discontinued. In a similar patient population, Ross et al. [bib_ref] Patients with chronic myeloid leukemia who maintain a complete molecular response after..., Ross [/bib_ref] used highly sensitive nested quantitative polymerase chain reaction (PCR) and found that patients who maintained a complete molecular response after stopping imatinib harbored a stable level of BCR-ABL. Taken together, these results suggest that imatinib therapy can induce durable remissions in a small subset of patients with CML and that the addition of IFNa may broaden imatinib's therapeutic potential in CML.
## Evaluating therapy with ifna
Kinetics and predictors of response to IFNa Patients respond more quickly to imatinib than to IFNa. Patients on IFNa therapy achieve CCyR at a median time of 19 months compared with 6 months on imatinib treatment. [bib_ref] for the European Study Group on Interferon in Chronic Myeloid Leukemia. Chronic..., Bonifazi [/bib_ref] [bib_ref] The achievement of an early complete cytogenetic response is a major determinant..., Jabbour [/bib_ref] This variation may be because of the different mechanisms of action of the two drugs.
Response to IFNa depends on the phase and duration of CML disease. In general, IFNa therapy best benefitted patients with early-stage disease and favorable prognostic factors. A low-risk prognostic profile included o1 year since diagnosis, no peripheral blood basophilia, no additional cytogenetic abnormalities, Caucasian descent and age o60 years; such patients achieved higher hematologic and cytogenetic response rates with IFNa than their high-risk counterparts. [bib_ref] Chronic myelogenous leukemia: a multivariate analysis of the associations of patient characteristics..., Kantarjian [/bib_ref] In addition, patients in accelerated phase and blast crisis did not typically respond to IFNa. IFNa also had limited effects in late CML-CP.
## Toxicities
Acute side effects to IFNa therapy commonly present as flu-like symptoms (anorexia, fever, chills, myalgias and headaches); these are not typically dose limiting and usually resolve in a few days. Chronic side effects include fatigue, weight loss, myalgias/ arthralgias, depression and immune-mediated complications, such as autoimmune hemolytic anemia/thrombocytopenia, collagen vascular disorders, hypothyroidism and immune-mediated nephritic syndrome. Cases of cardiac dysfunction, including dysrhythmias and congestive heart failure, are rare but require immediate discontinuation of IFNa. Chronic fatigue and neurotoxicity, such as depression and cognitive impairment, are common dose-limiting side effects and typically worsen with continued treatment. [bib_ref] Interferon in oncological practice: review of interferon biology, clinical applications, and toxicities, Jonasch [/bib_ref] As these toxicities have hindered compliance with therapy, three joint prospective studies examined whether a lower dose of IFNa at 3 MU/ m 2 five times a week would be as effective as the standard dose of 5 MU/m 2 daily. [bib_ref] Randomized comparison of low-dose versus high-dose interferon-alfa in chronic myeloid leukemia: prospective..., Kluin-Nelemans [/bib_ref] The studies found that overall survival and response rates did not dramatically differ between groups.
## Significance of cytogenetic responses to ifna
The critical finding of the IFNa trials was the correlation between cytogenetic response and survival. IFNa treatment led to MCyR in 10-40% of patients and CCyR in 5-30% of patients. [bib_ref] for the Italian Cooperative Study Group on Myeloid Leukemia. A randomized study..., Baccarani [/bib_ref] [bib_ref] Treatment of myelogenous leukemia: current status and investigational options, Kantarjian [/bib_ref] A group of European investigators created a registry of 317 patients with CML in CCyR after starting IFNa alone or with hydroxyurea. [bib_ref] for the European Study Group on Interferon in Chronic Myeloid Leukemia. Chronic..., Bonifazi [/bib_ref] The median time to first CCyR was 19 months. After 10 years, 72% of these patients were alive and 46% were in continuous CCyR. Similarly, Kantarjian et al. [bib_ref] Complete cytogenetic and molecular responses to interferon-alpha-based therapy for chronic myelogenous leukemia..., Kantarjian [/bib_ref] analyzed the long-term significance of cytogenetic responses to IFNa-based therapies. Of the 512 patients in early chronic phase, 27% achieved a CCyR within a median time of 16 months. These responders had a survival rate of 78% at a median follow-up of 127 þ months (range, 88 þ to 191 þ months). The induced CCyR was durable; patients who maintained cytogenetic remission for more than 2 years on IFNa therapy remained in remission for an average of 6 years after discontinuing treatment. These results, along with an additional study, [bib_ref] Follow-up of complete cytogenetic remission in patients with chronic myeloid leukemia after..., Mahon [/bib_ref] confirmed that CCyR predicts long-term survival in patients with CML and that IFNa can induce stable remissions in some patients with CML.
## Ifna mechanisms of action in cml
Although IFNa has been around for many years, we still do not know how it exerts its antileukemic effects. According to in vitro studies, IFNa modulates gene expression, promotes cell differentiation and apoptosis, directly inhibits cell growth and proliferation, restores regulation by the bone marrow microenvironment and induces an immunomodulatory response. Microarray analyses have shown that IFNa can induce expression of over 300 different genes. [bib_ref] Identification of genes differentially regulated by interferon a, b, or g using..., Der [/bib_ref] These genes encode apoptotic proteins (i.e., TRAIL, Fas, caspase-4, caspase-8 and XAF-1), anti-viral proteins (that is, PKR, 2 0 5 0 A oligoadenylate synthetase and Mx proteins), immunomodulatory proteins (that is, MHC I and II, LMP-2 and C1 inhibitor), host defense proteins (that is, PKR, IRF 1-9, interleukin-15 and interleukin-6) and transcription factors (that is, signal transducer and activator of transcription 1, signal transducer and activator of transcription 2, ISGF3-g and IRF1-7). [bib_ref] Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis, Chawla-Sarkar [/bib_ref] The precise function of many of the gene products induced by IFNa remains unknown; however, several of the identified genes encode well-known pro-apoptotic proteins, including TRAIL/Apo2L and Fas/CD95. [bib_ref] Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis, Chawla-Sarkar [/bib_ref] In CML progenitor cells, IFNa enhances the expression of the Fas receptor, thereby increasing cell sensitivity to Fas ligand. [bib_ref] Involvement of Fas-mediated apoptosis in the inhibitory effects of interferon-a in chronic..., Selleri [/bib_ref] In addition to activating apoptosis, IFNa directly targets key regulators of the cell cycle, including retinoblastoma protein, cdc25A, cyclins (cyclin D3, cyclin E and cyclin A) and cyclindependent kinases (cdk4 and cdk6). Such targeting can block and/ or lengthen the cell cycle phases, allowing cells to differentiate or undergo apoptosis. [bib_ref] Interferon a induces the expression of retinoblastoma gene product in human Burkitt..., Kumar [/bib_ref] [bib_ref] Type I interferon induction of the Cdk-inhibitor p21 WAF1 is accompanied by..., Subramaniam [/bib_ref] In bone marrow hematopoietic progenitors, IFNa directly inhibits proliferation by suppressing the production of hematopoietic stimulatory cytokines, such as granulocyte-macrophage colony-stimulating factor and interleukin-1b. It also increases the synthesis of inhibitory cytokines, including interleukin-1 receptor antagonist and transforming growth factor-b. [bib_ref] Influence of interferon-a on cytokine expression by the bone marrow microenvironment-impact on..., Peschel [/bib_ref] In addition, IFNa may inhibit the proliferation of CML progenitors by restoring normal hematopoietic mechanisms. In normal progenitors, b1-integrin receptors mediate cell adhesion to the bone marrow stroma, and stimulation of these receptors transmits antiproliferation signals. These two regulatory mechanisms are defective in CML progenitors, but Bhatia et al. [bib_ref] The effect of interferon-a on beta-1 integrin mediated adhesion and growth regulation..., Bhatia [/bib_ref] have shown that IFNa can restore them. Lastly, the growth-inhibitory effects of IFNa seem to require activation of the mitogen-activated protein kinase p38 in CML progenitors. [bib_ref] Signaling pathways activated by oncogenic forms of Abl tyrosine kinase, Zou [/bib_ref] [bib_ref] The p38 MAPK pathway mediates the growth inhibitory effects of interferon-a in..., Mayer [/bib_ref] IFNa treatment activates p38 by phosphorylation, which in turn leads to the transcription of IFNa-inducible genes.
In addition to directly inhibiting cell proliferation, IFNa may attenuate CML by activating host immune cells, including B and T lymphocytes, natural killer cells and antigen-presenting dendritic cells. [bib_ref] Effect of recombinant a-interferon on the expression of the bcr-abl fusion gene..., Andrews [/bib_ref] [bib_ref] Suppression of cell proliferation and the expression of a bcr-abl fusion gene..., Yanagisawa [/bib_ref] [bib_ref] Evidence that specific T lymphocytes may participate in the elimination of chronic..., Molldrem [/bib_ref] [bib_ref] Clinical reagents of GM-CSF and IFN-a induce the generation of functional chronic..., Weng [/bib_ref] The increased incidence of immune-mediated complications with IFNa therapy supports such immune activation. [bib_ref] Immunemediated and unusual complications during interferon alfa therapy in chronic myelogenous leukemia, Sacchi [/bib_ref] Addition of IFNa both in vitro and in vivo caused CML mononuclear cells to differentiate into dendritic cells; the dendritic cells then served as antigen-presenting cells for CMLspecific peptides. [bib_ref] Interferon-a induces dendritic cell differentiation of CML mononuclear cells in vitro and..., Paquette [/bib_ref] Similarly, in the presence of IFNa and granulocyte-macrophage colony-stimulating factor in vitro, CML bone marrow mononuclear cells differentiated into dendritic cells with specific antileukemia function. [bib_ref] Clinical reagents of GM-CSF and IFN-a induce the generation of functional chronic..., Weng [/bib_ref] This may have clinical relevance because addition of granulocyte-macrophage colonystimulating factor to treatment with IFNa in patients who failed to achieve an MCyR (n ¼ 14) improved cytogenetic responses for half of the patients. [bib_ref] GM-CSF can improve the cytogenetic response obtained with interferon-alpha therapy in patients..., Cortes [/bib_ref] Lastly, IFNa (but not imatinib) induces cytotoxic T cells (CTLs) specific for CML progenitors. [bib_ref] Interferon-a, but not the ABL-kinase inhibitor imatinib (STI571), induces expression of myeloblastin..., Burchert [/bib_ref] Given the potential benefits of combination therapy with imatinib and IFNa, a recent study investigated the effect of BCR-ABL signaling on IFNa activity in a CML cell line. [bib_ref] Bcr-abl signals to desensitize chronic myeloid leukemia cells to IFNa via accelerating..., Bhattacharya [/bib_ref] The study showed that expression of BCR-ABL in non-CML cells attenuated IFN signaling; however, pre-treatment of CML cells with imatinib augmented the antigrowth effects of IFNa exposure. In addition, imatinib pre-treatment enhanced signal transducer and activator of transcription 1 phosphorylation induced by IFNa. These results, in addition to providing insights into the mechanism of action of the combination therapy, may translate into a clinical strategy to increase the sensitivity of CML cells to IFNa.
## Using minimal residual disease to determine treatment plan
With the introduction of reverse transcriptase-polymerase chain reaction (RT-PCR) analysis, residual leukemic clones could be detected in patients thought to be in complete remission. In fact, RT-PCR is sensitive enough to detect one BCR-ABL þ cell per 1 Â 10 5 -1 Â 10 6 normal cells. [bib_ref] Detection of minimal residual bcr/abl transcripts by a modified polymerase chain reaction, Lee [/bib_ref] [bib_ref] Detection of two alternative bcr/abl mRNA junctions and minimal residual disease in..., Lee [/bib_ref] [bib_ref] Diagnosis of chronic myeloid and acute lymphocytic leukemias by detection of leukemiaspecific..., Kawasaki [/bib_ref] Lee et al. [bib_ref] Detection of minimal residual disease by polymerase chain reaction in Philadelphia chromosome-positive..., Lee [/bib_ref] demonstrated that all 29 patients with a CCyR to IFNa harbored some residual Ph þ cells. Even so, 21 of the patients maintained their CCyR at a median follow-up of 13 months after RT-PCR analysis. These findings suggested that PCR positivity for BCR-ABL does not predict immediate disease relapse. A clinical trial with longer follow-up of IFNa treatment revealed that cytogenetic remission can last for years, even when MRD resides in the early hematopoietic progenitor cells of patients with CML. [bib_ref] Persistence of dormant leukemic progenitors during interferon-induced remission in chronic myelogenous leukemia...., Talpaz [/bib_ref] One explanation is that IFNa puts tumor cells in a dormant state, which prevents residual leukemia cells from regenerating clinically significant leukemia.
Regardless of how remission is maintained in the presence of residual disease, the studies evaluating RT-PCR analysis raised the following two questions: (1) how long should patients continue IFNa therapy once they achieve a CCyR, and (2) can RT-PCR analysis guide the decision to discontinue therapy? To address these questions, Hochhaus et al. [bib_ref] for the German CML Study Group and the UK MRC CML Study..., Hochhaus [/bib_ref] continuously monitored BCR-ABL transcript levels by RT-PCR in 54 patients who were treated with IFNa and achieved CCyR. Over a median observation period of 1.9 years, the 14 patients who relapsed demonstrated a significantly higher median BCR-ABL:ABL ratio than those who maintained a CCyR (0.49% vs 0.021%; Po0.0001). These findings suggested that the degree of residual disease could predict the probability of relapse. When IFNa was withdrawn in six of the patients, one patient relapsed and was subsequently found to possess increasing levels of MRD; thus, the authors advised that IFNa be continued at least until low levels of BCR-ABL transcripts were achieved.
The IRIS trial was the first randomized trial to evaluate molecular disease by RT-PCR in patients with CML. The results showed that BCR-ABL transcript levels fell by 3 log or greater (defined as MMR) in 57% of patients with a CCyR after 12 months of imatinib treatment. In comparison, only 24% of patients with a CCyR in the IFNa group had at least a 3-log reduction in BCR-ABL transcripts. [bib_ref] for the International Randomised Study of Interferon versus STI571 (IRIS) Study Group...., Hughes [/bib_ref] Importantly, all patients who achieved an MMR remained progression free at the 24-month follow-up. A long-term followup of the IRIS trial examined patients who achieved a CCyR with imatinib (163 out of 553 patients at 18-month follow-up). [bib_ref] on behalf of the IRIS investigators. Long-term prognostic significance of early molecular..., Hughes [/bib_ref] Of these patients, 127 achieved an MMR by 18 months and none had CML progression at the 84-month follow-up. [bib_ref] on behalf of the IRIS investigators. Long-term prognostic significance of early molecular..., Hughes [/bib_ref] The authors concluded that once a CCyR was reached, RT-PCR assessment of molecular disease could replace cytogenetic analysis of patient response. In support of this, Kantarjian et al. [bib_ref] Complete cytogenetic and molecular responses to interferon-alpha-based therapy for chronic myelogenous leukemia..., Kantarjian [/bib_ref] reported that all 20 patients with persistent PCR-negative CML-CP maintained an MCyR at the last long-term follow-up 10 years from the first CCyR. Altogether, these studies validated molecular testing in CML and redefined how clinicians should measure patient responses and predict clinical outcomes.
To determine whether patients with sustained undetectable BCR-ABL transcript levels were fully cured or continued to generate leukemic stem cells in their bone marrow, a recent study was conducted in patients who achieved undetected MRD for 43 years with IFNa (n ¼ 3), imatinib after IFNa failure (n ¼ 2) or dasatinib after imatinib intolerance (n ¼ 1). [bib_ref] Leukemic stem cell persistence in chronic myeloid leukemia patients with sustained undetectable..., Chomel [/bib_ref] In all patients, Role of interferon in chronic myeloid leukemia M Talpaz et al leukemic stem cells expressing BCR-ABL were identified. At this time, whether undetected MRD correlates with risk of disease relapse is not known and warrants further investigation.
## Potential re-emergence of ifna use in cml: current developments and future strategies
Durable responses after discontinuation of IFNa Several cases of continuous cytogenetic remission after the cessation of IFNa therapy have generated excitement about the curative potential of IFNa. [bib_ref] Follow-up of complete cytogenetic remission in patients with chronic myeloid leukemia after..., Mahon [/bib_ref] [bib_ref] Long-term persistence of molecular response after discontinuation of interferon-alpha in two patients..., Veneri [/bib_ref] These patients usually maintained a CCyR for more than 24 months before discontinuing IFNa and maintained remission for an average of 6 years after discontinuing therapy. Outside of allogeneic hematopoietic stem cell transplantation, this represents the closest evidence of a clinical 'cure' for CML. Note that only approximately 20% of patients who receive IFNa achieve a durable CCyR, but those who reach this milestone demonstrate prolonged survival. Several preclinical studies provide possible reasons why imatinib may not be sufficient to cure CML. First, primitive CML cells/leukemic stem cells do not readily undergo apoptosis when exposed to imatinib, even after prolonged exposure. [bib_ref] Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are..., Graham [/bib_ref] [bib_ref] BMS-354825) targets an earlier progenitor population than imatinib in primary CML but..., Copland [/bib_ref] Second, CML early progenitor/stem cells persist in patients who achieve a CCyR with imatinib. [bib_ref] Persistence of malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete..., Bhatia [/bib_ref] Thus, since imatinib does not eliminate the malignant progenitors that cause the disease, it is probably not curative in the majority of cases. These progenitors may escape imatinib toxicity because they do not depend on BCR-ABL mechanisms for survival and proliferation. [bib_ref] The development of imatinib as a therapeutic agent for chronic myeloid leukemia, Deininger [/bib_ref] In support of this, a recent in vitro study demonstrated that while imatinib does inhibit the BCR-ABL kinase and its downstream signaling in CML primitive progenitors, the drug fails to facilitate death in these cells. [bib_ref] Human chronic myeloid leukemia stem cells are insensitive to imatinib despite inhibition..., Corbin [/bib_ref] This finding implies that CML stem cells are not 'addicted' to the BCR-ABL oncogene.
In contrast to therapy with imatinib, evidence suggests that IFNa actually targets the residual leukemic stem cells that cause disease relapse. Short-term colony-forming and long-term cultureinitiating cell assays showed that IFNa was more active against primitive CML progenitors, whereas imatinib preferentially targeted more mature, differentiated CML progenitors. [bib_ref] Effects of imatinib and interferon on primitive chronic myeloid leukaemia progenitors, Angstreich [/bib_ref] These findings may explain why the clinical responses to IFNa are slower but more durable than those to imatinib. Imatinib acts quickly on the more differentiated progenitors that make up the bulk of the leukemia. By contrast, since IFNa targets the rare CML stem cell (o1% of the CML population), its effects may not manifest as early on in treatment. More recent data in mice have shown that IFNa administered to dormant stem cells activates and thereby sensitizes them to subsequent killing by chemotherapeutic agents. [bib_ref] IFNa activates dormant haematopoietic stem cells in vivo, Essers [/bib_ref] A small clinical trial investigating the value of prior IFNa treatment in maintaining remission after discontinuing imatinib therapy is currently ongoing (ClinicalTrials.gov, Identifier: NCT01073436).
## Combining ifna with imatinib and second-generation tkis
To identify optimal imatinib-based regimens for CML-CP, two large multicenter, randomized treatment optimization studies were initiated: the German CML-Study IV (imatinib 400 mg vs imatinib plus IFNa (1.5-3 MU thrice weekly) vs imatinib plus cytosine arabinoside vs imatinib after IFNa failure vs imatinib 800 mg; n ¼ 1022) [bib_ref] Tolerability-adapted imatinib 800 mg/d versus 400 mg/d versus 400 mg/d plus interferon-a..., Hehlmann [/bib_ref] and the French STI571 Prospective Randomized Trial (SPIRIT; imatinib 400 mg vs imatinib 600 mg vs imatinib plus cytosine arabinoside vs imatinib plus PegIFNa; n ¼ 636). [bib_ref] Imatinib plus peginterferon alfa-2a in chronic myeloid leukemia, Preudhomme [/bib_ref] Of note, the German CML-Study IV used IFNa, whereas the French study used PegIFNa-2a (Pegasys; Hoffmann-La Roche Inc., Nutley, NJ, USA).
In the German CML-Study IV, recruitment to the imatinib plus cytosine arabinoside and the imatinib after IFNa failure study arms was terminated early because of feasibility and compliance issues. At 12 months, a higher rate of MMR was observed with tolerability-adapted imatinib 800 mg compared with the imatinib 400 mg ± IFNa arms (P ¼ 0.003; [fig_ref] Table 3: Efficacy data from CML-Study IV and SPIRIT Abbreviations [/fig_ref]. Significantly higher rates of CCyR were also observed with imatinib 800 mg over the first 24 months, as well as superior molecular responses at the 1, 0.1 and 0.01% BCR-ABL transcript levels according to the International Scale (IS). Treatment approaches were well tolerated with similar grade 3 and 4 adverse events. The investigators suggested that the superior remission rates in the high-dose imatinib arm were a result of the strategy applied (high dose early on and maintenance around 600 mg per day according to tolerability). The earlier and faster remissions with tolerability-adapted high-dose imatinib are expected to translate into better survival with longer follow-up.
In contrast to the German CML-Study IV, the French SPIRIT study reported significantly faster and better molecular response rates with the combination of imatinib plus PegIFNa-2a compared with imatinib alone (400 and 600 mg per day) and combined with cytarabine [fig_ref] Table 3: Efficacy data from CML-Study IV and SPIRIT Abbreviations [/fig_ref]. [bib_ref] Imatinib plus peginterferon alfa-2a in chronic myeloid leukemia, Preudhomme [/bib_ref] Specifically, the 12-, 18-and 24-month rates and cumulative incidences of major and superior (44-log reduction in BCR-ABL:ABL transcripts) molecular responses were significantly higher in this group. Enrollment in the imatinib 600 mg and imatinib plus cytosine arabinoside arms was stopped primarily because of low rates of molecular responses and observed toxicity, respectively. Furthermore, 45% of patients discontinued PegIFNa-2a in the first year primarily because of adverse effects; however, when the dose was reduced from 90 to 45 mg per week, treatment was better tolerated. A major finding of the study was that longer duration of imatinib plus PegIFNa-2a (particularly more than 12 months) correlated with a better rate of molecular responses. However, event-free survival did not differ across all the arms of the study after 4 years of follow-up. The second part of the trial will focus on whether the earlier and faster response rates with this combination translate into better survival.
At 12-month follow-up, the incidence of MMR with imatinib 400 mg per day was 31% in CML-Study IV and 38% in SPIRIT, both of which were comparable with the 39% MMR rate observed in IRIS. In contrast, the 12-month MMR rate in the imatinib plus IFNa arm of CML-Study IV was 35% compared with 57% in the imatinib plus PegIFNa-2a arm of SPIRIT. The patient populations in the two studies were not different, and it is possible that use of the pegylated form of IFNa, which was designed to have a longer half-life in the blood, improved the efficacy of the combination in the SPIRIT trial.
Three smaller phase II studies of imatinib plus PegIFNa-2b (PegIntron; Merck) reported discordant results. The Nordic group study (n ¼ 112) compared the combination of PegIFNa-2b 50 mg per week and imatinib 400 mg per day with imatinib 400 mg per day alone in patients with low-or intermediate-risk CML. [bib_ref] Almqvist A et al; for the Nordic CML Study Group. Combination of..., Simonsson [/bib_ref] The MMR rate was significantly higher in the combination arm (82%) compared with the monotherapy arm (54%) at 12 months. Even a short exposure to PegIFNa-2b (3-6 months) improved response to imatinib. Notably, 34 of the 56 patients in the combination arm discontinued PegIFNa-2b, mainly owing to adverse events, such as neutropenia and constitutional symptoms. To manage adverse events, the starting dose of PegIFNa-2b (50 mg per week) was lowered to 30 mg per week. Of those who continued therapy with PegIFNa-2b for more than 12 months, 91% achieved an MMR vs 58% in the imatinib monotherapy arm. The second two-arm study (n ¼ 94) examined the addition of PegIFNa-2b (0.5 mg/kg per week) and granulocyte-macrophage colony-stimulating factor to high-dose imatinib (800 mg per day) vs the continuation of high-dose imatinib alone in patients with early CML-CP. [bib_ref] Immune modulation of minimal residual disease in early chronic phase chronic myelogenous..., Cortes [/bib_ref] Unlike the Nordic and SPIRIT trials, this study found no differences in the cytogenetic or molecular response rates between the two arms. Owing to treatment-related toxicity, the combination arm had a high dropout rate, which may have hampered potential benefits of the immunotherapy. Adherence to PegIFNa-2b was also very low (13%) in an earlier study of imatinib plus PegIFNa-2b conducted by the Italian Cooperative Study Group. [bib_ref] on behalf of the GIMEMA Working Party on CML. Front-line treatment of..., Palandri [/bib_ref] [bib_ref] for the GIMEMA Working Party on Chronic Myeloid Leukemia. Imatinib and pegylated..., Baccarani [/bib_ref] The starting doses of PegIFNa-2b (50, 100 and 150 mg per week) were likely too high in combination with imatinib, leading to grade 3 or 4 neutropenia in 63% and grade 3 or 4 non-hematologic adverse events in 52% of patients. The German, SPIRIT and Nordic studies, which used lower doses of IFNa, reported comparatively fewer grade 3 and 4 hematological and non-hematological adverse events in patients taking imatinib plus IFNa. A retrospective analysis of the Italian study showed that CCyR and MMR rates were higher in patients receiving the combination at early time points, but were comparable with the imatinib-only arm at longer times. The durability of responses, event-free survival and overall survival were also similar between the two arms. [bib_ref] for the GIMEMA Working Party on Chronic Myeloid Leukemia. Imatinib and pegylated..., Baccarani [/bib_ref] [bib_ref] The response to imatinib and interferon-a is more rapid than the response..., Palandri [/bib_ref] Interestingly, the early and fast response rates with imatinib 800 mg and with imatinib plus PegIFNa are similar to those recently reported for the second-generation TKIs dasatinib [bib_ref] Dasatinib versus imatinib in newly diagnosed chronic-phase chronic myeloid leukemia, Kantarjian [/bib_ref] Value of IFNa in patients with the T315I mutation Treatment of patients with the T315I mutation (threonine-toisoleucine mutation at amino acid 315) in BCR-ABL is challenging, as this mutation confers resistance to treatment with imatinib, as well as second-generation TKIs. The frequency of this mutation ranges from 2-20% of imatinib-resistant CML patients. The National Comprehensive Cancer Network guidelines recommend hematopoietic stem cell transplantation, if applicable, or participation in a clinical trial for this patient population.Currently, a number of agents are under investigation, including ponatinib 105 (a reversible Abl-Src inhibitor), DCC-2036 (a switch-pocket inhibitor) and homoharringtonine (omacetaxine). [bib_ref] Current status of agents active against the T315I chronic myeloid leukemia phenotype, Burke [/bib_ref] Although no clinical studies have investigated the value of IFNa in treating patients with the T315I mutation, two case reports have recently been published. One patient with the T315I mutation achieved a CCyR after 12 months of treatment with imatinib (400 mg per day), but developed resistance after 18 months of treatment. The patient was treated with a combination of imatinib and IFNa (6 MU per week). After 51 months of combination treatment, he achieved MMR, and the T315I mutation was not detected by direct sequencing or pyrosequencing. While the patient experienced grade 2 anemia and grade 1 neutropenia and thrombocytopenia, he could continue the treatment with no dose reduction. [bib_ref] Successful treatment of a chronic-phase T-315I-mutated chronic myelogenous leukemia patient with a..., Itonaga [/bib_ref] In the second case, the patient was treated with KW-2449, a T315I-specific inhibitor, after losing the CCyR induced by 800 mg per day imatinib. While KW-2449 appeared to reverse the T315I mutation, his response did not improve; furthermore, he developed the F359I mutation. On switching to combination treatment with dasatinib (50 mg twice daily) and PegIFNa (9 MU per day), the patient achieved a CCyR and an MMR with a BCR-ABL:ABL ratio of 0.05 after 4 months. [bib_ref] Dasatinib combined with interferon-alfa induces a complete cytogenetic response and major molecular..., Cornelison [/bib_ref] These reports point to the potential value of IFNa in eradicating resistance to TKI treatment in the presence of the T315I mutation. More robust data are needed to confirm these treatment benefits.
Induction of CML-specific immunity by IFNa and implications for maintenance therapy As mentioned above, IFNa is known to activate leukemia-specific immunity, but the underlying mechanism is still not well understood. One current hypothesis is based on early work by [bib_ref] Evidence that specific T lymphocytes may participate in the elimination of chronic..., Molldrem [/bib_ref] Kanodia et al. [bib_ref] PR1-specific T cells are associated with unmaintained cytogenetic remission of chronic myelogenous..., Kanodia [/bib_ref] recently hypothesized that IFNa induces stable remissions at least in part by increasing the expression of PR1 in CML cells. As IFNa induces the expansion of self-renewing memory CTLs specific for PR1, the PR1-expressing CML progenitors become a prime target for immune-mediated killing. [bib_ref] PR1-specific T cells are associated with unmaintained cytogenetic remission of chronic myelogenous..., Kanodia [/bib_ref] In support of this, PR1-CTLs were found to be increased in CML patients with a CCyR after IFNa cessation. Moreover, the PR1-CTLs secreted IFNg in response to stimulation with PR1 peptide. By contrast, PR1-CTLs from the three patients who relapsed after IFNa withdrawal lost their ability to secrete IFNg. [bib_ref] PR1-specific T cells are associated with unmaintained cytogenetic remission of chronic myelogenous..., Kanodia [/bib_ref] These findings suggest that loss of functional PR1-CTLs may contribute to relapse in patients with CML. Given the potential for developing imatinib resistance and/or intolerance with continued imatinib treatment, maintenance therapy with IFNa may allow patients to discontinue imatinib by keeping CML progenitors suppressed. A recent report on the use of PegIFNa-2a maintenance after induction treatment with imatinib and PegIFNa-2a demonstrated sustained remission in 15 out of 20 CML-CP patients at a median of 2.4 years after imatinib discontinuation. [bib_ref] Sustained molecular response with interferon alfa maintenance after induction therapy with imatinib..., Burchert [/bib_ref] This impressive outcome was thought to involve a T-cell response because proteinase 3 mRNA levels and frequencies of PR1-CTLs increased during maintenance therapy with IFNa. To minimize toxicity from long-term IFNa use, a later study administered PegIFNa 9 months before and 3 months after imatinib discontinuation. [bib_ref] Treatment with interferon alpha prior to discontinuation of imatinib in patients with..., Hardan [/bib_ref] This regimen improved the remission status of 5 of the 11 patients over a median follow-up of 47 months. These studies support further exploration of the role of IFNa consolidation or maintenance therapy after TKI induction.
OPTIMIZING THERAPY: EARLY RESPONSE PREDICTORS Before the imatinib era, the Hasford or Euro score (developed from a study of 1303 IFNa-treated patients) was used to predict prognosis at diagnosis based on spleen size, percent blasts, age, platelet count, eosinophilia and basophilia. [bib_ref] Writing Committee for the Collaborative CML Prognostic Factors Project Group. A new..., Hasford [/bib_ref] A new prognostic score called EUTOS (European Treatment and Outcome Study score) has since been developed to predict clinical responses to imatinib. [bib_ref] Predicting complete cytogenetic response and subsequent progression-free survival in 2060 patients with..., Hasford [/bib_ref] The score was developed from a study of 2060 patients treated with imatinib, including imatinib at 800 mg and in combination with IFNa. Using only two variables (spleen size and basophil percentage in peripheral blood), the score discriminates between high-and low-risk groups and predicts that 34% of highrisk patients will fail to achieve CCyR. This score predicts treatment failure with better sensitivity and specificity than the Sokal or Euro scores. [bib_ref] Predicting complete cytogenetic response and subsequent progression-free survival in 2060 patients with..., Hasford [/bib_ref] A further advance is response prediction at 3 months. CML patients at risk of progression are candidates for change of therapy, including addition of IFNa to front-line TKI treatments. Identifying patient response to a drug early on in the treatment is key for optimizing treatment protocols. [bib_ref] Harmonization of molecular monitoring of CML therapy in Europe, Mü Ller [/bib_ref] Early response predictors for CML are summarized in . Patients who achieve a cytogenetic remission (CCyR or MCyR) or reach a BCR-ABL level of o10% (IS) after 3 months have significantly better overall survival after 5 years (95% vs 87%). [bib_ref] The achievement of an early complete cytogenetic response is a major determinant..., Jabbour [/bib_ref] [bib_ref] Early molecular and cytogenetic response is predictive for long-term progression-free and overall..., Hanfstein [/bib_ref] [bib_ref] Assessment of BCR-ABL1 transcript levels at 3 months is the only requirement..., Marin [/bib_ref] This will likely replace the current definition of optimal response to imatinib at 3 months, which requires CHR and o95% Ph þ metaphases. [bib_ref] Chronic myeloid leukemia: an update of concepts and management recommendations of European..., Baccarani [/bib_ref] For newly diagnosed patients treated with IFNa, achievement of CHR within 3 months predicted MCyR. [bib_ref] Response at three months is a good predictive factor for newly diagnosed..., Mahon [/bib_ref]
# Conclusions
The reduced tolerability and slower response kinetics of IFNa compared with TKIs have reduced the enthusiasm for this therapy. However, therapy with imatinib and other TKIs may be limited by drug resistance, intolerance and, when therapy is discontinued, relapse. In contrast to targeted therapies, IFNa has a broad range of therapeutic effects that may reduce the likelihood of resistance or relapse, especially when used in combination with other CML therapies. These factors along with the proven efficacy of pegylated forms of IFNa, even at low doses, have revived interest in IFNa therapy for CML.
## Conflict of interest
JC is a consultant for and has received research funding from BMS, Novartis, Ariad and Chemgenex. MT has chaired a satellite symposium for Merck and has received drugs from Merck for clinical studies. The remaining authors declare no conflict of interest.
[table] Table 2: Combination trials of IFNa: a historical overview [/table]
[table] Table 1: Single-agent trials of IFNa: a historical overview [/table]
[table] Table 3: Efficacy data from CML-Study IV and SPIRIT Abbreviations: CML, chronic myeloid leukemia; NA, not available; PegIFNa, pegylated IFNa; SPIRIT, STI571 Prospective Randomized Trial.Molldrem et al.109 examining the immunogenicity of proteinase 3, a serine protease highly expressed in various myeloid leukemia cells, including CML. A peptide derived from proteinase 3 known as PR1 was identified with high affinity for HLA-A.2.1. Of significance, CTLs specific for PR1 eliminated CML progenitors, but not normal marrow cells. A subsequent investigation detected circulating PR1-specific CD8 þ T cells in 11 out of 12 IFNa responders, but not in non-responders (0 of 7). [/table]
|
Near-Roadway Pollution and Childhood Asthma: Implications for Developing “Win–Win” Compact Urban Development and Clean Vehicle Strategies
Near-Roadway Pollution and Childhood Asthma: Implications for Developing "Win-Win" Compact Urban Development and Clean Vehicle StrategiesNear-Roadway Pollution and Childhood Asthma: Implications for Developing "Win-Win" Compact Urban Development and Clean Vehicle Strategies
1 Online supplement
Supplemental Material, : Yearly number of childhood asthma-related exacerbations attributable to dispersion-modeled near-roadway pollution in combination with reduction of regional NO 2 (top) and regional O 3 (bottom) above background levels in clean communities (scenario 1, dispersion-modeled NOx model) (95% confidence intervals in parentheses) a .
Exacerbations due to regional air pollution among children with asthma caused by… |
A subxiphoid uniportal video-assisted thoracoscopic surgery for synchronous bilateral pulmonary metastasis: A case report
[bib_ref] A subxiphoid single-incision video-assisted thoracoscopic lobectomy: A case report, Tezel [/bib_ref] [bib_ref] Singleincision subxiphoid approach for bilateral metastasectomy, Suda [/bib_ref] [bib_ref] A subxiphoid single-incision video-assisted thoracoscopic lobectomy: A case report, Tezel [/bib_ref] [bib_ref] Singleincision subxiphoid approach for bilateral metastasectomy, Suda [/bib_ref]
## Case report
A 39-year-old female patient was admitted to our clinic with nodular lesions in the bilateral lung. Her medical history revealed that she was operated on the right knee region about three years ago due to an osteosarcoma. Thoracic computed tomography (CT) showed 1x1-cm nodules in the right lung lower lobe basal segment and the left lung upper lobe posterior segment [fig_ref] Figure 1: Computed tomography images showing bilateral pulmonary calcific and solid nodules [/fig_ref]. However, no other nodules were detected in the lung parenchyma, even in millimeters. Also, the nodules contained calcified areas. We performed whole body positron emission tomography (PET) screening for distant metastases; however, no other fluorodeoxyglucose (FDG) involvement site was seen. Pulmonary nodules were planned to be excised for diagnostic and metastasectomy based on the recommendation of the Oncological Council. Since there were no thoracic CT findings showing adhesion between the lung and the pleura and the patient did not previously undergo any thoracic surgery, we planned to perform metastasectomy in the same session in both lungs. A written informed consent was obtained from the patient.
Bilateral wedge resection was planned synchronously with uniportal subxiphoid video-assisted thoracoscopic surgery (VATS). To perform wedge resection, the patient was ventilated by performing left selective intubation with a double-lumen intubation tube. Following the endotracheal intubation, a 4-cm incision was made in the subxiphoid in the position. The operating table was, then, placed at 30° right lateral decubitus position. Exploration was performed in the left hemithorax with 5-mm 30° video thoracoscopic optics. Since there was a single and peripheral nodule in both lungs, it was not difficult to detect the localization of the lesion and, therefore, it was not necessary to use carbon dioxide (CO 2 ) insufflation during the operation. The nodule in the left lung upper lobe posterior segment was suspended with a grasper and wedge resection was performed using two endostaplers [fig_ref] Figure 2: Intraoperative images of pulmonary solid nodules with VATS [/fig_ref]. After bleeding and air leak control, 20-Fr thoracic drains were placed in the thorax. Since no other nodules were seen in the left lung and pleura of the patient, excision of the nodule in the right lung was started. The patient was placed 30° left lateral decubitus position. The VATS camera was entered into the thorax, and wedge resection was applied to the nodule in the lower lobe basal segment of the right lung [fig_ref] Figure 2: Intraoperative images of pulmonary solid nodules with VATS [/fig_ref]. To prevent tumor implantation in the incision area, wedge resected nodules were removed out of the thorax using an endobag. The procedure was terminated by placing a 20-Fr drain into the thorax. Both thoracic drains were inserted through the incision line in the subxiphoid area. The drain on the left side was removed on the third postoperative day, and the drain on the right side was removed on Day 4. The patient was discharged on the postoperative Day 4. In the postoperative follow-up in the outpatient setting, chest radiography revealed normal findings. The pathological examination result was reported as an osteosarcoma and the patient was referred to oncology clinic for the treatment.
# Discussion
Lung metastases of sarcoma species are common and metastasectomy is known to improve survival in these patients. [bib_ref] Assessment of metastasectomy and prognostic factors in the treatment of metastatic lung..., Şengül [/bib_ref] It has been reported that the rate of sarcomas among tumors treated with pulmonary metastasectomy is about 18%. [bib_ref] Results of resection of pulmonary metastasis and prognostic factors, Cangel [/bib_ref] As a surgical technique, thoracotomy, VATS or median sternotomy are utilized. However, it is important to choose less invasive technique in cases in whom the number and location of nodules are apparent on thoracic CT. In the studies of Cangel et al., [bib_ref] Results of resection of pulmonary metastasis and prognostic factors, Cangel [/bib_ref] the results of 118 surgical interventions were reported, and VATS was applied effectively and safely in eligible cases.
Less pain and faster recovery can be achieved by increasingly using the options of VATS such as uniportal and subxiphoid incision in the practice of thoracic surgery. With the incision applied from the subxiphoid area, the surgical procedure can be easily performed without damaging the intercostal nerve. In addition, since it is possible to reach both hemithoraces from this region, surgery can be performed bilaterally in both lungs in the same session. Bilateral surgery has been shown to be effective, particularly in the presence of a limited number of nodules in the lung and in lesions with a peripheral location. It can be also considered as an advantage that the patient does not receive anesthesia twice in a certain period of time unlike surgeries performed in the right and left separate sessions. [bib_ref] Uniportal subxiphoid video-assisted thoracoscopic bilateral segmentectomy for synchronous bilateral lung adenocarcinomas, Aresu [/bib_ref] With uniportal subxiphoid VATS, the feasibility of all resection options in the lung has been demonstrated. This makes it to be a preferred technique among thoracic surgeons. It has also positive aspects such as reduced pain, favorable cosmetic appearance, early recovery, and less paresthesia in the chest wall. However, it is thought to have a disadvantage in terms of bleeding control in the major bleeding which may occur during lung major resections. Carvalheiro et al. [bib_ref] Uniportal VATS Lobectomy: Subxiphoid Approach, Carvalheiro [/bib_ref] reported that mediastinal lymph node sampling was insufficient, especially when performing lung cancer surgery; however, the same study recommends the subxiphoid approach because of its advantages.
Postoperative pain increases, as the number of ports increases in VATS. With the subxiphoid uniportal VATS, less pain is perceived, as there is no intercostal nerve compression. Therefore, we believe that it should be preferred in suitable cases. As in our case, subxiphoid incision can be used easily in patients with a low number of nodules located peripherally.
In conclusion, subxiphoid uniportal video-assisted thoracoscopic surgery has certain advantages such as excellent pain control, quick recovery, shorter hospital stay, and synchronous bilateral lung intervention. Based on these findings, we can speculate that this technique would be more widely adopted in the near future among thoracic surgeons.
[fig] ©2020: All right reserved by the Turkish Society of Cardiovascular Surgery. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes (http://creativecommons.org/licenses/by-nc/4.0/). [/fig]
[fig] Figure 1: Computed tomography images showing bilateral pulmonary calcific and solid nodules; (a, b) left lung upper lobe posterior segment (arrow), (c, d) right lung lower lobe basal segment (arrowhead). [/fig]
[fig] Figure 2: Intraoperative images of pulmonary solid nodules with VATS. (a) Right lung peripheral nodule (arrow), (b) left lung nodule (arrowhead). VATS: Video-assisted thoracoscopic surgery. [/fig]
|
Immunological Predictors of Dimethyl Fumarate‐Induced Lymphopenia
View this article online at wileyonlinelibrary.com.
Treatment with dimethyl fumarate (DMF) leads to lymphopenia and infectious complications in a subset of patients with multiple sclerosis (MS). Here, we aimed to reveal immune markers of DMF-associated lymphopenia. This prospective observational study longitudinally assessed 31 individuals with MS by single-cell mass cytometry before and after 12 and 48 weeks of DMF therapy. Employing a neural network-based representation learning approach, we identified a CCR4-expressing T helper cell population negatively associated with relevant lymphopenia. CCR4-expressing T helper cells represent a candidate prognostic biomarker for the development of relevant lymphopenia in patients undergoing DMF treatment. D imethyl fumarate (DMF) is an oral immunomodulatory compound approved as first-line treatment for relapsing-remitting multiple sclerosis (MS). [bib_ref] Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis, Gold [/bib_ref] [bib_ref] Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis, Fox [/bib_ref] [bib_ref] Dimethyl fumarate: a review in relapsing-remitting MS, Blair [/bib_ref] [bib_ref] Update on treatment in multiple sclerosis, Callegari [/bib_ref] Unlike more targeted disease-modifying treatments like monoclonal antibodies, DMF affects various leukocyte populations including T cells, and myeloid cells. [bib_ref] Dimethyl fumarate influences innate and adaptive immunity in multiple sclerosis, Diebold [/bib_ref] [bib_ref] Therapeutic efficacy of dimethyl fumarate in relapsing-remitting multiple sclerosis associates with ROS..., Carlström [/bib_ref] Lymphopenia is the most important safety concern of this treatment. Patients with relevant lymphopenia, specifically at lymphocyte counts < 700/μl, experience an increased risk of developing progressive multifocal leukoencephalopathy (PML) caused by a reactivation of the JC polyomavirus (JCV). [bib_ref] Classifying PML risk with disease modifying therapies, Berger [/bib_ref] [bib_ref] Progressive multifocal leukoencephalopathy in dimethyl fumarate-treated multiple sclerosis patients, Jordan [/bib_ref] To assess this risk under therapy, serial lymphocyte counts and age are used in clinical routine. The early identification of patients at risk of developing lymphopenia is therefore of crucial relevance to guide clinicians in their therapy choice.
Here, we developed a deep immunophenotyping approach based on high-dimensional mass cytometry in conjunction with a weakly supervised machine learning algorithm [bib_ref] Sensitive detection of rare disease-associated cell subsets via representation learning, Arvaniti [/bib_ref] to identify predictive markers of lymphopenia in peripheral blood.
# Materials and methods
## Study design
We prospectively collected blood samples from a clinically wellcharacterized cohort of 31 MS patients intending to start DMF treatment (for baseline characteristics, see . Peripheral blood mononuclear cells (PBMCs) from these patients were sampled at baseline, and 3 and 12 months after initiation of DMF therapy (T1, T2, and T3, respectively). Participants were grouped by their lowest lymphocyte counts during 12-month follow-up into 2 groups: patients with (n = 10) and without (n = 21) lymphopenia (lymphocyte counts < 700/μl; for study design, see [fig_ref] FIGURE 1: Mass cytometric analysis of the immune profile of dimethyl fumarate [/fig_ref]. The study was approved by the Ethics Committee for Northwest and Central Switzerland.
## Sample processing and cryopreservation
Blood samples were characterized by automated flow cytometry at the time point of collection for main leukocyte and lymphocyte populations (including CD3/CD4, CD3/CD8, CD19, and CD56 cells). PBMCs were collected as previously described and cryopreserved in liquid nitrogen. [bib_ref] Dimethyl fumarate influences innate and adaptive immunity in multiple sclerosis, Diebold [/bib_ref] Short-term reactivation and stimulation of cryopreserved PBMCs were performed as described recently. [bib_ref] GM-CSF and CXCR4 define a T helper cell signature in multiple sclerosis, Galli [/bib_ref]
## Mass cytometry
PBMCs were analyzed by a standardized mass cytometry procotol with live cell barcoding for a panel of 36 markers, containing lineage/activation markers, chemokine receptors, and intracellular cytokines. After standardized preprocessing, data were subjected to algorithm-based high-dimensional analysis as described recently. [bib_ref] GM-CSF and CXCR4 define a T helper cell signature in multiple sclerosis, Galli [/bib_ref]
# Statistical analysis
Unless otherwise specified, results indicate 2-tailed t tests with Benjamini-Hochberg correction for multiple testing performed using R or Prism (GraphPad Software, San Diego, CA). Probability values of <0.05 were considered significant. Pearson correlation coefficients (r) were derived from the z statistic of the Mann-Whitney-Wilcoxon test. For mass cytometry data, preprocessed datasets were randomly downsampled to 300,000 cells per donor. Samples were clustered with FlowSOM and annotated according to their protein-expression patterns. Uniform Manifold Approximation and Projection (UMAP) visualization used a reduced dataset of 10,000 randomly selected cells per patient (plotted using ggplot2). CellCNN 14 is a weakly supervised machine learning model used to detect and define phenotype-associated (here, lymphopenia-associated) rare cell subpopulations. The algorithm was trained to identify rare cell populations that explain the difference between phenotypes. We implemented the algorithm in a longitudinal design including all time points in a single model. To this end, time residuals T3-T1 and T3-T2 were correlated with relevant lymphopenia, using a 3-fold cross-validation. For more technical details on CellCNN, we refer to Arvaniti and Claassen. [bib_ref] Sensitive detection of rare disease-associated cell subsets via representation learning, Arvaniti [/bib_ref]
# Results
We assessed a cohort of 31 DMF-treated individuals with MS, of whom 10 developed relevant lymphopenia with lymphocyte counts < 700/μl during the follow-up period (see [fig_ref] FIGURE 1: Mass cytometric analysis of the immune profile of dimethyl fumarate [/fig_ref] for clinical characteristics, . First, we characterized the effect of DMF therapy on the general composition of PBMCs. We used the unsupervised representation learning algorithm FlowSOM 16 to delineate the major immune populations and map their relationships in a UMAP representation. We found a strong effect of DMF treatment on the T cell compartment, with a significant reduction in the frequency of CD8, CD4, and γδ T . This effect was more pronounced in patients developing lymphopenia. Conversely, relative myeloid and B cell frequencies were only minimally affected in nonlymphopenic individuals, whereas they increased in most patients with relevant lymphopenia. Overall, the immune pattern under treatment revealed an underrepresentation of T cells in lymphopenic patients.
To determine whether lymphopenia could be predicted based on the homeostatic immune profile of patients before DMF therapy, we implemented the supervised representation-learning algorithm CellCNN 14 to identify cellular features able to stratify patients developing lymphopenia through the longitudinal assessment of feature changes from baseline to the 3-and 12-month time points. With this approach, the highest predictive accuracy of 73.96% was achieved for an effector memory T helper cell phenotype [fig_ref] FIGURE 2: CellCNN identifies a cellular signature predicting lymphopenia development [/fig_ref]. The decrease in this lymphopeniaassociated population was significant at all 3 time points including baseline in individuals with lymphopenia, indicating a predisposition rather than a specific treatment effect of DMF. This cell cluster was constituted almost exclusively of CD4+ T cells with an effector memory phenotype. Phenotypically, this population is characterized by high expression of CCR4, CD25, CD103, and IL10.
To test the strength of our findings, we compared this newly identified biomarker with other known immunological and clinical lymphopenia-associated parameters in a multivariate regression analysis. We confirmed previous findings that older age and female sex influence the risk of developing lymphopenia. In this model, however, only the frequency of CellCNN-identified effector memory T helper cell phenotype reached an independent significance level [fig_ref] FIGURE 3: CellCNN-identified signature predicts lymphopenia development in dimethyl fumarate [/fig_ref]. In conclusion, these results suggest that-independently from age and overall lymphocyte counts-CCR4-and IL10-expressing CD4 cells act as a predictor of relevant lymphopenia under DMF treatment.
# Discussion
Treatment-associated lymphopenia and its infectious complication PML constitute a major concern in DMF-treated patients. So far, we lack reliable biomarkers to predict this potentially life-threatening condition. Employing highdimensional immune profiling and a machine learning algorithm enabled us to assess circulating immune cells in unprecedented detail. This analysis of the mass-cytometric dataset not only confirmed the fundamental rearrangement of most subsets of both T cells [bib_ref] Reduction of CD8(+) T lymphocytes in multiple sclerosis patients treated with dimethyl..., Spencer [/bib_ref] [bib_ref] Dimethyl fumarate influences innate and adaptive immunity in multiple sclerosis, Diebold [/bib_ref] [bib_ref] Dimethyl fumarate treatment alters circulating T helper cell subsets in multiple sclerosis, Gross [/bib_ref] [bib_ref] Dimethyl fumarate induces a persistent change in the composition of the innate..., Diaz [/bib_ref] [bib_ref] Effects of dimethyl fumarate on lymphocyte subsets, Berkovich [/bib_ref] and B cells, [bib_ref] Dimethyl fumarate treatment mediates an anti-inflammatory shift in B cell subsets of..., Li [/bib_ref] [bib_ref] Dimethyl fumarate treatment of relapsing-remitting multiple sclerosis influences B-cell subsets, Lundy [/bib_ref] described in the existing literature, but identified an immune cell signature predictive of lymphopenia.
This population is largely composed of effector memory T helper cells sharing conventional regulatory T cell features such as the production of the cytokine IL10, and high expression of CD25 and CD103, as well as chemokine receptor CCR4. In contrast to putatively diseasepropagating effector memory T cell clusters (identified by assessment of different treatment approaches 19 or crosssectional comparison of diseases [bib_ref] GM-CSF and CXCR4 define a T helper cell signature in multiple sclerosis, Galli [/bib_ref] , the population described here is characterized by relatively low expression of VLA4, IL17a, GM-CSF, and CXCR4. [bib_ref] GM-CSF and CXCR4 define a T helper cell signature in multiple sclerosis, Galli [/bib_ref] [bib_ref] Cellular immunology of relapsing multiple sclerosis: interactions, checks, and balances, Bar-Or [/bib_ref] [bib_ref] Identifying CNScolonizing T cells as potential therapeutic targets to prevent progression of..., Kaufmann [/bib_ref] Because pathogenic T helper cells are known to be drastically decreased by DMF treatment, [bib_ref] GM-CSF and CXCR4 define a T helper cell signature in multiple sclerosis, Galli [/bib_ref] we hypothesize that the residual mature T cell pool is decisive for the formation of relevant lymphopenia. Our newly described lymphopenia-associated population can be interpreted as a refined indicator of this residual T cell pool and may allow estimation of the individual capacity for T cell reduction before treatment start.
The central role of memory T helper cells in the pathophysiology of PML 20 draws further attention to this potential indicator population. Low levels and slow increase of central-memory and effector memory T helper cells were described as an immunological core feature of fatal PML under DMF treatment. [bib_ref] Low frequencies of central memory CD4 T cells in progressive multifocal leukoencephalopathy, Dubois [/bib_ref] Avoiding cell loss in this immune cell compartment-where not therapeutically necessary-therefore appears to be a desirable goal for prevention of opportunistic infections. As previously reported, higher age and lower lymphocyte counts at baseline are already clinically established for risk stratification of relevant lymphopenia and PML occurrence, [bib_ref] Classifying PML risk with disease modifying therapies, Berger [/bib_ref] but the relevance of the composition of the T cell compartment has not yet been analyzed. The predictive effector memory population described may help to substantially refine this approach in two aspects. First, it represents an additional, independent, and significant tool only loosely correlated with patient age and the overall lymphocyte counts. Furthermore, the here-described T helper cell population-in conjunction with pathogenic T cells in MS-could play a relevant role in the immunological control of latent JCV. Although these observations require corroboration in larger cohorts, the data available suggest that screening of effector memory T helper cells-with high CCR4 and IL10 expression-before start of DMF therapy may offer a prospective risk stratification tool.
[fig] FIGURE 1: Mass cytometric analysis of the immune profile of dimethyl fumarate (DMF)-treated patients. Peripheral blood mononuclear cells (PBMCs) of DMF-treated patients were longitudinally collected and analyzed through mass cytometry. (A) Schematic description of the analyzed cohort. Lymphopenia has been defined by <700 lymphocytes/μl in laboratory testing. (B) Neural network-guided definition of immune cell lineages. Mean population expression levels of all markers used for Uniform Manifold Approximation and Projection (UMAP) visualization and FlowSOM clustering. (C) The Uniform Manifold Approximation and Projection algorithm (1,000 cells, randomly selected from each individual patient at 3 different time points [n = 93]) was used to depict different populations therein. FlowSOM-based immune cell populations are overlaid as a color dimension. (D) Frequencies of immune cell lineages in peripheral leukocytes of multiple sclerosis patients developing lymphopenia (n = 10) and not developing lymphopenia (n = 21) with DMF therapy.cells [/fig]
[fig] FIGURE 2: CellCNN identifies a cellular signature predicting lymphopenia development. (A) CellCNN-selected signature cells (colored) are overlaid on a UMAP visualization of the major immune cell lineages from all samples. (B) Relative frequencies of the CellCNN-identified population at 3 different time points stratified by clinical groups. (C). Frequency of selected cell types within the lymphopenia-associated population (LAP) at T1 in the conventional panel. The color code is identical with Figure 1C. (D) Expression patterns of the 5 key discriminant markers between the LAP and the reference cell population for the stimulated panel. Distance between patterns for each marker is quantified by Kolmogorov-Smirnov (KS) test. (E) Heatmap depicting the mean expression level of clustering markers used to define different memory subsets in T helper cells. T helper cells from all donors and from all 3 time points were used. (F) Frequency of different T helper memory clusters within CellCNN-selected signature cells. Each time point is represented. (G) Mean population expression levels of analyzed parameters in T helper compartment and in the lymphopenia-associated cell signature. PBMC = peripheral blood mononuclear cell; Tcm = T central memory; Tem = T effector memory; TEMRA = terminal effector.678Volume 91, No. 5 [/fig]
[fig] FIGURE 3: CellCNN-identified signature predicts lymphopenia development in dimethyl fumarate (DMF)-treated patients. (A) Regression modeling of main predictors of lymphopenia and (B) graphical representation of normalized estimates. Probability values are based on 2-tailed Mann-Whitney-Wilcoxon tests between multiple sclerosis patients developing lymphopenia (n = 10) and not developing lymphopenia (n = 21) with DMF therapy. *p < 0.05. EDSS = Expanded Disability Status Scale; M = male. [/fig]
|
Research Progress on the Cardiac Injury from ACE2 Targeting in SARS-CoV-2 Infection
# Introduction
In December 2019, pneumonia outbreaks with unexpected results occurred throughout the world. Scientists subsequently identified a novel coronavirus, which lifted the veil of this infectious viral pneumonia. This novel coronavirus pneumonia caused widespread concern throughout the world. In February 2020, the International Committee on Taxonomy of Viruses gave this coronavirus an official name, "SARS-CoV-2." At the same time, the World Health Organization (WHO) named it "coronavirus disease 2019 (COVID-19)". This epidemic has now spread to all parts of the world. As of 18 January 2021, the confirmed total number of COVID-19 cases reached 93,805,612 globally, and the total number of deaths was 2,026,093. These figures continue to rise each day and bring a serious threat to human health, social and economic development, and the global medical and public health system. Scientists around the world are actively developing effective treatment strategies, focusing on vaccines and antiviral agents. The most frequently used antiviral agents include chloroquine, hydroxychloroquine, and remdesivir. Among them, the more in-depth studies are focusing on remdesivir.
Remdesivir belongs to inhibitors of viral RNA polymerase/RNA synthesis; it has a broad spectrum antiviral activity against several viruses. Replication of SARS-CoV-2 requires the viral RNA-dependent RNA polymerase (RdRp) enzyme, a target of remdesivir. It has shown in vitro activity against SARS-CoV-2. Remdesivir appears to have a favorable clinical safety profile. SARS-CoV-2 can enter the body through glycoprotein recognition, and this process can be blocked by vaccines. When the vaccines act on the body, they can induce the body to produce neutralizing antibodies targeting glycoprotein and block the virus from entering the host cells. Previously, a variety of vaccines had been
## The basic structure of sars-cov-2
On 10 January 2020, the SARS-CoV-2 full genome sequence was published after metagenomic RNA sequencing. SARS-CoV-2 is a single-stranded RNA virus with a total genome length of approximately 29,903 nucleotides and constituting ten genes. Its specific genome composition includes untranslated regions (UTR) at both ends of the RNA strand and a complete open reading frame (ORF) that can encode 9860 amino acids. The RNA strand of the SARS-CoV-2 genome includes a methylated "cap" at the 5 end and a poly-A "tail" structure at the 3 end. ORFs at the 5 end can encode a series of viral replicases, of which ORF1b and ORF1a encode 16 nonstructural proteins. The four ORFs at the 3 end encode four structural proteins, which are the spike (S), the membrane (M), the nucleocapsid (N), and the envelope (E) proteins, respectively. The S protein functions critically in mediating the virus's adsorption and fusion with the host cell membrane. Wrapp et al. employed cryo-electron microscopy and 3D reconstruction techniques to obtain the S protein's trimeric structure at a resolution of 3.5 Å. The S protein belongs to trimeric class I fusion protein, which contains two functional subunits, S1 and S2. S1 binds to host cell receptors via its RBD, while S2 fuses the viral membrane with the host cell membrane, which allows the viral genome to enter the host cell. This entrance is an extremely complicated procedure that requires the synergy of receptor binding and proteolysis. The basic structure of SARS-CoV-2 forms the structural foundation for cardiac injury via the targeting of ACE2 during the invasion. The basic structure of SARS-CoV-2 is summarized in.
Biomolecules 2021, 11, x FOR PEER REVIEW 3 of 13. Wrapp et al. employed cryo-electron microscopy and 3D reconstruction techniques to obtain the S protein's trimeric structure at a resolution of 3.5 Å. The S protein belongs to trimeric class I fusion protein, which contains two functional subunits, S1 and S2. S1 binds to host cell receptors via its RBD, while S2 fuses the viral membrane with the host cell membrane, which allows the viral genome to enter the host cell. This entrance is an extremely complicated procedure that requires the synergy of receptor binding and proteolysis. The basic structure of SARS-CoV-2 forms the structural foundation for cardiac injury via the targeting of ACE2 during the invasion. The basic structure of SARS-CoV-2 is summarized in. . The RNA strand of the SARS-CoV-2 genome has a methylated "cap" at the 5′ end, and a poly-A "tail" structure at the 3′ end. ORF1a and ORF1b at the 5′ end can encode 16 nonstructural proteins. The four ORFs at the 3′ end encode four structural proteins, namely, the spike (S), the envelope (E), the membrane (M), and the nucleocapsid (N) proteins. The S protein consists of two functional subunits, S1 and S2.
## The basic structure and physiological role of ace2
On ACE2 is a metalloproteinase with a total length of 805 amino acids, including a 17amino-acid N-terminal signal peptide (one zinc-binding motif) and a C-terminal membrane anchor. A unique collectrin domain is also included. Although ACE2 and ACE are homologs and are both parts of the RAS, their actions are opposite. ACE can convert angiotensin (Ang) I into Ang II, which can activate the G protein-coupled receptor angiotensin II type 1 receptor (AT1R) to promote vasoconstriction, increase vascular permeability, and mediate inflammatory responses. ACE2 can cleave Ang I and produce Ang-(1-9) peptide that can be transformed into vasodilator peptide Ang-(1-7) through ACE and other peptidases. In addition, ACE2 can hydrolyze Ang II into Ang-(1-7). In turn, the latter can act on Mas receptors to achieve vasodilation; reduce vascular permeability; exert antiproliferative, antioxidant, and antithrombotic effects; and reverse myocardial hypertrophy, left ventricular remodeling, and myocardial fibrosis. Therefore, it exerts essential functions in the cardiovascular system. In the human body's normal physiological state, the ACE2/Ang-(1-7)/MAS and ACE/Ang II/AT1R axes are in a state of dynamic equilibrium, thereby maintaining the normal functioning of the heart.
The cardiovascular system appears to have complex interactions with COVID-19. Poor clinical outcomes are associated with COVID-19 patients who have hypertension, coronary heart disease, or cardiac injury. A vicious circle between SARS-CoV-2 infection and cardiac dysfunction or heart failure may occur. The putative relationship between them may relate to the role of ACE2. ACE2 is a key element in the renin-angiotensin-aldosterone system (RAAS), which systemically affects the vasculature and blood pressure. SARS-CoV-2 virus enters human cells via binding its surface "spike" to ACE2. This interaction was significantly enhanced in patients with hypertension, coronary heart disease, diabetes, or other comorbidities. Within the heart, cardiomyocytes, endothelial, and pericytes all express ACE2. Since ACE2 plays an important role in . The RNA strand of the SARS-CoV-2 genome has a methylated "cap" at the 5 end, and a poly-A "tail" structure at the 3 end. ORF1a and ORF1b at the 5 end can encode 16 nonstructural proteins. The four ORFs at the 3 end encode four structural proteins, namely, the spike (S), the envelope (E), the membrane (M), and the nucleocapsid (N) proteins. The S protein consists of two functional subunits, S1 and S2.
## The basic structure and physiological role of ace2
On ACE2 is a metalloproteinase with a total length of 805 amino acids, including a 17-amino-acid N-terminal signal peptide (one zinc-binding motif) and a C-terminal membrane anchor. A unique collectrin domain is also included. Although ACE2 and ACE are homologs and are both parts of the RAS, their actions are opposite. ACE can convert angiotensin (Ang) I into Ang II, which can activate the G protein-coupled receptor angiotensin II type 1 receptor (AT1R) to promote vasoconstriction, increase vascular permeability, and mediate inflammatory responses. ACE2 can cleave Ang I and produce Ang-(1-9) peptide that can be transformed into vasodilator peptide Ang-(1-7) through ACE and other peptidases. In addition, ACE2 can hydrolyze Ang II into Ang-(1-7). In turn, the latter can act on Mas receptors to achieve vasodilation; reduce vascular permeability; exert antiproliferative, antioxidant, and antithrombotic effects; and reverse myocardial hypertrophy, left ventricular remodeling, and myocardial fibrosis. Therefore, it exerts essential functions in the cardiovascular system. In the human body's normal physiological state, the ACE2/Ang-(1-7)/MAS and ACE/Ang II/AT1R axes are in a state of dynamic equilibrium, thereby maintaining the normal functioning of the heart.
The cardiovascular system appears to have complex interactions with COVID-19. Poor clinical outcomes are associated with COVID-19 patients who have hypertension, coronary heart disease, or cardiac injury. A vicious circle between SARS-CoV-2 infection and cardiac dysfunction or heart failure may occur. The putative relationship between them may relate to the role of ACE2. ACE2 is a key element in the renin-angiotensin-aldosterone system (RAAS), which systemically affects the vasculature and blood pressure. SARS-CoV-2 virus enters human cells via binding its surface "spike" to ACE2. This interaction was significantly enhanced in patients with hypertension, coronary heart disease, diabetes, or other comorbidities. Within the heart, cardiomyocytes, endothelial, and pericytes all express ACE2. Since ACE2 plays an important role in SARS-CoV-2 infection, the high expression of ACE2 in the heart increases the risk of SARS-CoV-2 infection.
A cardio-protective effect of ACE2 has been reported in various animal models and clinical studies. Previous studies have consistently shown that the function of ACE2 is lost when SARS-CoV-2 binds to ACE2, which is mainly due to endocytosis and activation of proteolysis. ACE2 expression was significantly decreased and myocardial dysfunction occurred in mice after human SARS-CoV infection, indicating that ACE2 plays a key role in mediating SARS-CoV infection in the heart. If ACE2 does protect heart function in SARS-CoV-2 patients, the loss of ACE2 will further aggravate the burden on the heart, which forms a vicious circle. A lot of evidence indicates that the cardiovascular system is involved in the severity of disease infected by COVID-19, but the specific potential mechanism is still unclear. Further clinical trials are needed to clarify the role of ACE2 in SARS-CoV-2 patients and the detailed molecular mechanism of cardiac injury and heart failure.
## Structural characteristics causing cardiac injury
Similarities in the structures and gene sequences between the S proteins of SARS-CoV-2 and SARS-CoV suggest that they might share a receptor, i.e., the ACE2. Subsequent studies confirmed that similar to SARS-CoV, the SARS-CoV-2 S protein mediates host cell entry, with ACE2 as the invasion target. Compared with SARS-CoV, the structural conformation regarding the SARS-CoV-2 S protein interaction with ACE2 has been preserved overall, despite some changes in four essential amino acids in five binding sites regarding the SARS-CoV-2 S protein to the human ACE2. The SARS-CoV-2 binding affinity is 10-20 times stronger than that of SARS-CoV, explaining the enhanced humanto-human SARS-CoV-2 transmission. In addition, a study by Wan's team further confirmed that residue 394 (glutamine) of the RBD of the SARS-CoV-2 S protein corresponds to residue 479 in SARS-CoV, which could be identified via the key lysine 31 residue of the human ACE2 receptor. This finding suggests that SARS-CoV-2 may recognize human ACE2 more effectively compared with SARS-CoV.
Furthermore, ACE2 exists as a dimer in the human body and exhibits both open and closed conformations. These two conformations contain mutual recognition sites for coronaviruses, thus providing SARS-CoV-2 favorable conditions to infect humans with ACE2 as the target. The distribution of ACE2 in the human body is organ-and cell-specific. Using single-cell RNA sequencing, Zou et al. revealed that ACE2 is highly expressed in type II alveolar epithelial cells and cardiomyocytes, which indicates that the heart is also an important organ involved in SARS-CoV-2 infection.
## The possible mechanism of cardiac injury following the targeting of ace2 after viral invasion
It has been reported that different levels of cardiac injury occurred in some hospitalized patients with COVID-19. However, the mechanism regarding cardiac injury due to SARS-CoV-2 infection is not entirely elucidated. Based on published literature and previous studies on SARS-CoV, two possible mechanisms are proposed: direct and indirect injury. The direct injury mechanism refers to the cardiac injury caused by SARS-CoV-2 resulting from the targeting of ACE2 upon invasion. The indirect injury mechanism refers to the cardiac injury caused by cytokine storms following the host's immune responses. Previous investigations show that cytokine storm after SARS-CoV-2 infection may also result from targeting ACE2.
Since SARS-CoV-2 virus enters human cells via binding its surface "spike" to ACE2, the high expression of ACE2 in the cardiovascular system may directly accelerate the attack of SARS-CoV-2 virus. Some studies have reported that ACE2 can exert myocardial protective effects, reduce myocardial fibrosis, and improve cardiac function. in vivo experiments on mice and found they can develop ACE2dependent myocardial infection. In addition, autopsies of SARS patients have also documented reductions in ACE2 mRNA and protein expression levels in cardiomyocytes, with macrophage infiltration and frank myocardial damage. Moreover, the upregulation of ACE2 has been found after using RAAS inhibitors such as ACE inhibitors and angiotensin receptor type 1 blockers (ARBs), which is more likely to increase the patient's susceptibility to SARS-CoV-2 or show a worse prognosis. To the contrary, other research revealed that it would not increase the risk of susceptibility. Previous Biomolecules 2021, studies have shown that the function of ACE2 is lost when SARS-CoV-2 binds to ACE2, which counteracts the cardioprotective effect of ACE2. Besides, this will create a vicious circle, aggravating heart damage. Further investigation revealed that the patient's serum Ang II levels were significantly elevated. This finding may be because once the human body is infected with SARS-CoV-2, ACE2 will be depleted, which will affect the dynamic equilibrium between ACE2/Ang-(1-7)/MAS and ACE/Ang II/AT1R axes, eventually leading to impaired cardiac function. The direct injury mechanism of SARS-CoV-2 using ACE2 as a target, leading to cardiac injury, is summarized in.
also documented reductions in ACE2 mRNA and protein expression levels in cardiomyocytes, with macrophage infiltration and frank myocardial damage. Moreover, the upregulation of ACE2 has been found after using RAAS inhibitors such as ACE inhibitors and angiotensin receptor type 1 blockers (ARBs), which is more likely to increase the patient's susceptibility to SARS-CoV-2 or show a worse prognosis. To the contrary, other research revealed that it would not increase the risk of susceptibility. Previous studies have shown that the function of ACE2 is lost when SARS-CoV-2 binds to ACE2, which counteracts the cardioprotective effect of ACE2. Besides, this will create a vicious circle, aggravating heart damage. Further investigation revealed that the patient's serum Ang II levels were significantly elevated. This finding may be because once the human body is infected with SARS-CoV-2, ACE2 will be depleted, which will affect the dynamic equilibrium between ACE2/Ang-(1-7)/MAS and ACE/Ang II/AT1R axes, eventually leading to impaired cardiac function. The direct injury mechanism of SARS-CoV-2 using ACE2 as a target, leading to cardiac injury, is summarized in. The direct mechanism of cardiac injury caused by SARS-CoV-2 using ACE2 as the target. Angiotensinogen is converted to Ang I by renin. Angiotensin-converting enzyme (ACE) can convert Ang I into Ang II, which in turn can activate the angiotensin II type 1 receptor (AT1R). ACE2 can cleave Ang I to produce the Ang-(1-9) peptide, which can then be converted into the vasodilator peptide Ang-(1-7) through ACE or other peptidases. Conversely, ACE2 can hydrolyze Ang II into Ang-(1-7), which acts on Mas receptors. When the SARS-CoV-2 S protein binds to ACE2 receptors, the dynamic equilibrium is destroyed between the ACE/Ang II/AT1R and ACE2/Ang-(1-7)/MAS axes, leading to impaired cardiac function.
In addition, the indirect injury mechanism is also essential. Cytokine storm caused by immune disorder may be a key mediator. The hypothesis is that the overactivated immune system is one of the causes of SARS-CoV-2 induced heart injury. Sriramula et al. have shown that in rat models of Ang II-induced hypertension, a significant increase in inflammatory cytokines such as interleukin 6 (IL-6), interleukin 1β (IL-1β), and tumor necrosis factor α (TNF-α) occurs. In contrast, after the enhanced expression of ACE2, there. The direct mechanism of cardiac injury caused by SARS-CoV-2 using ACE2 as the target. Angiotensinogen is converted to Ang I by renin. Angiotensin-converting enzyme (ACE) can convert Ang I into Ang II, which in turn can activate the angiotensin II type 1 receptor (AT1R). ACE2 can cleave Ang I to produce the Ang-(1-9) peptide, which can then be converted into the vasodilator peptide Ang-(1-7) through ACE or other peptidases. Conversely, ACE2 can hydrolyze Ang II into Ang-(1-7), which acts on Mas receptors. When the SARS-CoV-2 S protein binds to ACE2 receptors, the dynamic equilibrium is destroyed between the ACE/Ang II/AT1R and ACE2/Ang-(1-7)/MAS axes, leading to impaired cardiac function.
In addition, the indirect injury mechanism is also essential. Cytokine storm caused by immune disorder may be a key mediator. The hypothesis is that the overactivated immune system is one of the causes of SARS-CoV-2 induced heart injury. Sriramula et al. have shown that in rat models of Ang II-induced hypertension, a significant increase in inflammatory cytokines such as interleukin 6 (IL-6), interleukin 1β (IL-1β), and tumor necrosis factor α (TNF-α) occurs. In contrast, after the enhanced expression of ACE2, there was a decrease in the relevant inflammatory cytokines, and the hypertension was controlled. Haga et al. found that after SARS-CoV infection there was a reduction in ACE2 and a significant rise in inflammatory cytokines, including TNF-α and IL-1β. These two studies suggest that the release of inflammatory cytokines may be related to ACE2. In senescent mice that were infected with SARS-CoV, CD4+ T cells can produce neutralizing antibodies and balance the immune response. Along with exhaustion of CD4+ and CD8+ T cells, severe immune system disorders appear in SARS-CoV-2 patients. Another retrospective study of 69 COVID-19 patients found that patients showed elevated hs-TNI, CK-MB, IL-6, C-reactive protein, and procalcitonin, whereas their lymphocyte count and CD4+/CD8+ ratio were reduced. In a case analysis published by Huang et al., COVID-19 patients showed elevated IL-1β, interferon γ (IFN-γ), monocyte chemoattractant protein 1 (MCP-1), and interferon-induced protein 10 (IP-10). Critically ill intensive care unit (ICU)patients showed more pronounced increases in IL-7, IL-2, IL-10, granulocyte-macrophage colonystimulating factor (GM-CSF), MCP-1, IP-10, and TNFα compared with patients with mild disease. These findings indicate that cytokine storms are related to the severity of viral infection.
NLRP3 inflammasome, the powerful proinflammatory system, may contribute to the initiation of a cytokine storm in the development of SARS-CoV-2 pathologies. Proinflammatory stimuli such as cell debris can induce the expression of NLRP3 and other inflammatory body components in cardiomyocytes. In fact, persistent virus shedding occurs in the death cases of COVID-19 infection. These virus fragments act heterologously to further activate the immune system and eventually lead to a cytokine storm. SARS-CoV-2 spike protein may directly trigger the activation of enzyme activity and downstream signal transduction after binding to ACE2 expressed on cell surface or release some effective cleavage fragments like C3a and C5a, which can directly trigger the activation of NLRP3 inflammatory bodies. Inflammation of the vascular system can lead to diffuse microvascular disease and thrombosis and then affect the function of the cardiovascular system.
In summary, SARS-CoV-2 infection is commonly followed by an ACE2 decrease, RAS imbalance, Ang II elevation, and a sharp increase in proinflammatory cytokines, thereby directly or indirectly damaging cardiomyocytes. Direct and indirect injury are not completely independent in clinic. Direct injury can lead to overactivation of the inflammatory response, which in turn aggravates the damage of the cardiovascular system.
## Manifestations of cardiac injury
Researchers in China and abroad have performed autopsies and puncture histopathology on patients who died of SARS-CoV-2 infection. They found degeneration and necrosis in cardiomyocytes; a small amount of monocyte, lymphocyte, and (or) neutrophil infiltration in the interstitium; and endothelial cell shedding, intimal inflammation, and thrombosis in a portion of blood vessels. In terms of clinical symptoms, Huang et al. were the first to publish a study involving 41 patients. Among the 41 COVID-19 patients in their study, five patients had acute myocardial injuries, manifesting elevated levels of serum hypersensitive troponin I (hs-cTHI) (>28 pg/mL), four of whom were critically ill and admitted to ICU for treatment. Subsequently, another team analyzed 84 cases of COVID-19 and found that some patients showed an increase in their myocardial enzyme spectrum, especially in creatine kinase (CK) and creatine kinase-myocardial band (CK-MB), suggesting the presence of myocardial injury. In a clinical analysis of COVID-19 diagnosed by healthcare workers, five of the 30 confirmed cases showed concomitant myocardial injury. Furthermore, among the 138 COVID-19 cases confirmed and enrolled in Wang et al., ten patients presented cardiac injuries. Among them, ICU patients had significantly higher CK-MB and hs-cThI levels than non-ICU patients. Similarly, among these 138 confirmed cases of COVID-19, 23 cases showed arrhythmia, but further analysis could not be performed as the specific type of arrhythmia was unknown, which indicates the necessity of electrocardiogram monitoring in clinical work. Furthermore, among the 99 confirmed cases of COVID-19 published by Wuhan Hospital in China, 11 patients died of sudden cardiac death, and these patients did not have a history of cardiovascular disease. Peng et al. reported another 112 confirmed cases of COVID-19. Among them, 27.6% had a concomitant cardiovascular disease and a higher case fatality rate.
There are some phenotypes of cardiac injury induced by COVID-19 infection. Myocardial injury has been reported in 20-30% of hospitalized patients with COVID-19 infection. When combined with basic diseases such as hypertension, diabetes, and coronary heart disease, the risk of death will increase. Microvascular and coagulation dysfunction Biomolecules 2021, are also manifestations of cardiac injury. Multiple reports confirmed that the elevated plasma D-dimer and fibrin degradation products that occurred were accompanied by prolonged prothrombin time or thrombocytopenia in COVID-19 infection patients. Diffuse microvascular disease and thrombosis may lead to heart and other organ infarction and further worsen multiple organ failure. Arrhythmia is also a phenotype of heart injury. In a report of 138 hospitalized COVID-19 patients, arrhythmias occurred in 16.7% of patients. Inflammation and cytokine storm may also reflect the manifestations of cardiac injury, it may be a phenotype or a cause of further aggravation of cardiac injury as discussed above.
In conclusion, SARS-CoV-2 infection is likely to result in cardiac injury, and the degree of cardiac injury is positively related to disease severity. Therefore, clinical symptoms should be closely observed to provide early treatment and reduce mortality.
## Therapeutic strategies using ace2 as a potential target
There are currently no effective and specific therapeutic strategies for SARS-CoV-2 treatment. Many treatment options may be derived by summarizing the treatment experiences for SARS and MERS. The primary research and development (R&D) strategy is the screening of existing broad-spectrum antiviral drugs and further developing drugs that can simultaneously target the virus and the host. Previous studies have found that SARS-CoV-2 may invade host cells via ACE2 receptors, which has provided many targets for the R&D of treatments and SARS-CoV-2 vaccines, including the blockade of the SARS-CoV-2 S protein binding with ACE2 receptors via the application of ACE inhibitors, among other strategies.
Several glycosylation sites can be found in the extracellular structure of ACE2 expressed in mammalian cells. The glycosylation of these sites may affect SARS-CoV-2 S protein binding with the ACE2 receptor, providing us with new strategies to block SARS-CoV-2 S protein binding with the ACE2 receptor. Additionally, recent studies have discovered that chloroquine can block viral infection through the improvement of endosomal pH required for virus/cell fusion and interfere with ACE2 terminal glycosylation. Furthermore, China has listed chloroquine phosphate as a therapeutic drug for COVID-19 patients in the "Diagnosis and Treatment Protocol for Coronavirus Disease 2019 (Trial Version 7)" and has included this drug in large-scale clinical trials. Several medical teams in China and abroad have developed ACE2 antibodies, S protein antibodies, and others based on the blockade of binding between the ACE2 receptor and the SARS-CoV-2 S protein. At present, there is still much controversy surrounding the use of ACE inhibitor-like drugs in patients with cardiac injury after SARS-CoV-2 infection. There are two main types of ACE inhibitor-like drugs that have received significant attention: angiotensin-converting enzyme inhibitors (ACEI) and angiotensin II receptor blockers (ARB). SARS-CoV-2 infection can lead to decreased ACE2, an imbalance between the ratio of ACE and ACE2, an absolute or relative increase in Ang II, and an over-activation of AT1R, which can result in impaired cardiac function. ACEI and ARB may inhibit the above pathophysiological changes and improve cardiac injury after SARS-CoV-2 infection.
Meanwhile, some reports revealed that the use of common classes of antihypertensive medications did not increase the risk of positive or severe COVID-19. These data support the continuation of existing treatment for patients with hypertension during the COVID-19 pandemic. There are also some contrary views: some studies revealed that although administering hypertensive rats with ACEI/ARB can reduce their blood pressure, their ACE2 levels were elevated by 4.7-and 2.8-fold, respectively. This means the application of ACEI/ARB may increase ACE2 negative feedback and increase infection risk. Liu et al. argued that the application of ACE inhibitor-like drugs might increase cells' susceptibility to viral invasion or worsen the disease. This supposition is because ACE inhibitors can lead to an increase in bradykinin levels, which can lead to vasodilation and lower blood pressure, as well as causing edema and exacerbating the inflammatory response.
In summary, the application of ACE inhibitor-like drugs after SARS-CoV-2 infection, especially in patients with cardiac injury, still requires support from a large amount of clinical data to validate these conclusions.
# Discussion
The epidemic caused by SARS-CoV-2 infection has achieved global focus. The WHO has also designated this viral epidemic as a significant global public health emergency. However, thus far, our understanding of SARS-CoV-2 is only the "tip of the iceberg." Many researchers have reported that the SARS-CoV-2 sequence identity is at most 88% homologous with bat-derived coronaviruses, suggesting that bats may be a natural host. Subsequently, researchers from the South China University of Technology found through the analysis of more than 1000 metagenomic samples that pangolins may also be a candidate intermediate SARS-CoV-2 host. The primary sources of SARS-CoV-2 infection are persons infected with this virus. The main transmission routes are via respiratory droplets and contact transmission, with high population susceptibility. Concerning SARS-CoV-2 infection pathogenesis, current studies were combined with the existing literature to conclude that SARS-CoV-2 invades host cells via the mediation of the S protein targeting ACE2. Many experimental studies and reports of clinical symptoms in patients with COVID-19 suggest that the heart is a potential target organ for SARS-CoV-2 infection. Moreover, the mechanism of cardiac injury may be caused directly by ACE2 depletion resulting from SARS-CoV-2 binding with ACE2 and indirectly by a cytokine storm. It is hoped that the findings of the above studies will provide new directions for the future development of ACE2 as a therapeutic target for cardiac injury after SARS-CoV-2 infection, as well as for related vaccines.
Based on a large number of reported clinical cases of SARS-CoV-2, we found that in addition to cardiac and pulmonary injury, SARS-CoV-2 infection might also lead to kidney, liver, gastrointestinal tract, testicular, and even ocular injury. This damage may be due to the organ and cell specificity of ACE2 distribution, such that SARS-CoV-2 can attack multiple organs using ACE2 as a target. It may also be related to the immune defense mechanisms, whereby the intensity of the immune response varies from individual to individual, and some studies have shown that autoimmune attack may also cause multiple organ injuries. This response will be a key focus in future research. The mechanism of cardiac injury after SARS-CoV-2 infection remains unclear. Apart from the mechanism involving ACE2 as the target of invasion, which is highlighted in this review, there may be many other mechanisms involved, such as the series of pathophysiological changes induced by hypoxemia, leading to cardiac injury. Disorders of pulmonary gas exchange in patients with COVID-19 may lead to hypoxemia, while acidosis, oxidative stress, and induced inflammatory reactions during hypoxia and reperfusion can exacerbate cardiac injury. Finally, among patients infected with SARS-CoV-2, 50% have chronic underlying diseases, such as hypertension, heart disease, and diabetes. These patients are more prone to develop severe diseases. During the treatment of these patients, SARS-CoV-2 infection was only the initial illness, and the final causes of death are often heart failure and multiple organ dysfunction. Therefore, we should be more vigilant when it comes to the cardiac condition of COVID-19 patients in clinical practice, complying with the principles of individualization in the application and selection of ACE inhibitor-like drugs for these patients.
Of course, safe and effective vaccines are also urgently needed. As we know, it is not feasible for COVID-19 to obtain group immunity through human infection. In the development process of new crown vaccines, researchers are facing a variety of problems and challenges, such as the weak immunogenicity of a single dose vaccine, incomplete virus inactivation, disease risk, and safe mass production. However, we believe that with the development of effective vaccines and antiviral drugs we will eventually defeat SARS-CoV-2 infection.
## Institutional review board statement: not applicable.
Informed Consent Statement: Not applicable.
# Data availability statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.
## Conflicts of interest:
The authors declare no conflict of interest. |
Roll eccentricity extraction and compensation based on MPSO-WTD and ITD
To meet the high thickness accuracy requirements in cold-rolling processes, a roll eccentricity signal extraction method based on modified particle swarm optimization and wavelet threshold denoising (MPSO-WTD) with intrinsic time-scale decomposition (ITD) is proposed. The strong denoising ability of the wavelet is combined with the decomposition and recognition attributes of ITD for non-stationary signals. Periodic disturbances in strip thickness caused by roll eccentricity are actively compensated. First, the wavelet is used to denoise the signal and the MPSO algorithm is applied to determine a rational threshold and improve the calculation efficiency. Then, the denoised signal is decomposed into proper rotational components (PRCs) using the ITD method, and an appropriate PRC component representing the eccentricity signal is extracted. Finally, the eccentricity compensation signal is applied in the automatic gauge control (AGC) system of the cold rolling mill. During the rolling process, the rolling speed is not constant and will directly affect the frequency of the roll eccentricity signal. To solve this problem, an encoder is installed at the end of the roll and the compensation frequency of the roller eccentricity signal is determined in the roller eccentricity compensation system according to the pulse number output. The results of simulations and experiments show that roll eccentricity signals extracted using the proposed method can effectively remove the influence of interference signals. An average improvement of 62.3% in the roll eccentricity compensation effect was achieved under the stable rolling condition in the finishing rolling stage. OPEN ACCESS Citation: Gao S, Xu L, Li Y, Ji J (2022) Roll eccentricity extraction and compensation based on MPSO-WTD and ITD. PLoS ONE 17(2): e0259810.
# Introduction
In strip production, aluminum alloy strip quality is one of the most important factors affecting consumer selection when deciding among similar competing products. Strip quality is mainly assessed by thickness error and flatness error. In the rolling process, many factors can affect strip quality by causing deviation in thickness or defects, and roll eccentricity is a key factor [bib_ref] Online Monitoring System Design for Roll Eccentricity in Rolling Mills, Xu [/bib_ref]. Roll eccentricity often exists in strip rolling processes, and is generally caused by inexact roll grinding, work roll, or back-up roll ovality [bib_ref] Active compensation of roll eccentricity in rolling mills, Kugi [/bib_ref]. The existence of roll eccentricity can lead to periodic fluctuations in rolling force and roll gap and may adversely affect the control effects of traditional automatic gauge control (AGC) systems [bib_ref] Roll eccentricity extraction and compensation based on improved wavelet denoising and EEMD, Li-Xin Wei [/bib_ref]. Thus, roll eccentricity compensation is essential. Accurate compensation relies on accurately extracting the eccentricity signal, which can be affected by a variety of disturbance signals.
At present, there are a variety of methods for analyzing roll eccentricity signals, such as neural network prediction methods [bib_ref] Coiling eccentricity compensation control system based on BP neural network algorithm, Sun [/bib_ref] [bib_ref] Strip Thickness Control of Cold Rolling Mill with Roll Eccentricity Compensation by..., Hameed Waleed [/bib_ref] [bib_ref] On-line identification of roll eccentricity based on PSO-RBF neural network, Fengji [/bib_ref] [bib_ref] Neural network for identification of roll eccentricity in rolling mills, Aistleitner [/bib_ref] , fast Fourier transform (FFT) and modified FFT (MFFT) algorithms [bib_ref] A method for analysis and compensation of roll eccentricity signal based on..., Lixin [/bib_ref] [bib_ref] FFT algorithm sampled with hanning-window for roll eccentricity analysis and compensation, Yong [/bib_ref] [bib_ref] Prognosis system for roll eccentricity with MFFT based on the difference evolution..., Wang [/bib_ref] , and wavelet transforms. A neural network-based technique was previously presented for identifying three key factors of roll eccentricity based on the measured angular velocity of rolls [bib_ref] Neural network for identification of roll eccentricity in rolling mills, Aistleitner [/bib_ref]. However, only the fundamental wave can be identified leading to limited compensation accuracy. Rolling mill stands without angular velocity sensors are common due to cost constrains and economic considerations [bib_ref] Online Monitoring System Design for Roll Eccentricity in Rolling Mills, Xu [/bib_ref]. Furthermore, large amounts of data are required to train the models, which places high demands on hardware systems resulting in low calculation efficiency [bib_ref] Roll eccentricity extraction and compensation based on improved wavelet denoising and EEMD, Li-Xin Wei [/bib_ref]. The Fourier transform is widely used to process linear stationary signals and can be used to effectively analyze the frequency characteristics of signals. However, the method is not suitable for processing local signal information and introduces distortion in denoising nonlinear and non-stationary signals [bib_ref] An EEMD-SVD-LWT algorithm for denoising a lidar signal, Xiao [/bib_ref] [bib_ref] Improvement of detection performance on single photon lidar by EMD-based denoising method, Wu [/bib_ref]. Although the difference evolution algorithm used in the MFFT removes some restrictions of the traditional Fourier transform computational efficiency is reduced [bib_ref] Online Monitoring System Design for Roll Eccentricity in Rolling Mills, Xu [/bib_ref].
In contrast to the FFT, wavelet transform analysis offers good localization characteristics in both the time and frequency domains [bib_ref] Fault diagnosis in industrial induction machines through discrete wavelet transform, Bouzida [/bib_ref]. Moreover, the wavelet transform is not restricted by sampling duration requirements or influenced by signal acquisition noise, and has therefore been widely used in denoising operations. The wavelet threshold denoising method can conveniently and flexibly extract roll eccentricity signals. However, it should be noted that frequency aliasing and redundant images can emerge during wavelet decomposition and reconstruction process [bib_ref] Roll eccentricity compensation based on anti-aliasing wavelet analysis method, Zhiming [/bib_ref] , therefore, the method does not guarantee that roll eccentricity components will reflects the real situation in rolling mills.
Many scholars have attempted to overcome the frequency aliasing phenomenon in wavelet decomposition using demodulation methods or combining wavelet denoising with various algorithms. Demodulation methods include generalized demodulation signal decomposition [bib_ref] Application of the improved generalized demodulation time-frequency analysis method to multi-component signal..., Cheng [/bib_ref] , iterative generalized demodulation [bib_ref] Time-frequency analysis of time-varying modulated signals based on improved energy separation by..., Feng [/bib_ref] , and parameter resolution modulation [bib_ref] Component extraction for non-stationary multi-component signal using parameterized de-chirping and band-pass filter, Yang [/bib_ref]. Frequency demodulation methods can eliminate cross interference among signal components under certain conditions, however, instantaneously intersecting frequency signal components cannot be separated. An eccentricity signal extraction method combining improved wavelet denoising and ensemble empirical mode decomposition (EEMD) was previously proposed [bib_ref] Roll eccentricity extraction and compensation based on improved wavelet denoising and EEMD, Li-Xin Wei [/bib_ref]. The EEMD method can suppress the frequency aliasing phenomenon of the wavelet algorithm and improve the eccentric signal extraction accuracy. However, when dealing with time-varying non-stationary signals, the number of signal components obtained by EEMD is usually larger than the actual number of characteristic components, therefore, false components with no correlation to the signal characteristics may arise, and calculation efficiency is low [bib_ref] Review of Signal Decomposition Theory and Its Applications in Machine Fault Diagnosis, Chen [/bib_ref].
The intrinsic time-scale decomposition (ITD) method can overcome spectral aliasing and has high computational efficiency and precision, however, its noise resistance is poor [bib_ref] Review of Signal Decomposition Theory and Its Applications in Machine Fault Diagnosis, Chen [/bib_ref]. At present, the ITD method is mainly used in the field of mechanical fault diagnosis for diagnosing gear faults [bib_ref] Fault Diagnosis of Gearbox based on ITD-Tunable Q-Factor Wavelet Transform, Verma Jay Govind [/bib_ref] and diesel engine faults [bib_ref] Application of complete ensemble intrinsic time-scale decomposition and least-square SVM optimized using..., Jun-Hong [/bib_ref]. Previous research has demonstrated high accuracy of the ITD method in signal feature extraction.
Roll eccentricity is not constant in rolling process. For instance, the threading situation as well as speed up/down situation, the rolling speed varies quickly, so as to the change of eccentricity frequency, and the eccentricity amplitude will also vary due to the abrasion of rolls [bib_ref] Online Monitoring System Design for Roll Eccentricity in Rolling Mills, Xu [/bib_ref]. To improve the accuracy of roller eccentricity signal compensation, the influence of rolling speed and roll wear on eccentricity signal should be considered.
To improve roll eccentricity signal extraction, this paper proposes an MPSO-WTD method with ITD. The algorithm combines the advantages of wavelet analysis with those of the ITD method. To prevent changes in rolling speed from influencing the roller eccentricity signal compensation system, an encoder is installed at the end of the roll and the compensation frequency of the roller eccentricity signal is controlled according to the pulse number output. Roll wear accumulates slowly and can influence product yield. To further ensure stability of the control system, the eccentricity compensation signal is periodically modified online according to the set rolling production requirement.
Finally, to verify the effect of the eccentricity compensation signal on improving strip thickness characteristics, compensation signals were input into the AGC system of a four-high irreversible cold strip rolling mill.
# Mpso-wtd method
## Wavelet threshold function
Denoising methods using wavelet thresholding are based on the assumption that the energy of the useful part of the signal will be concentrated in a small number of large-amplitude coefficients. However, most noise energy is dispersed throughout a large number of small-amplitude coefficients. Based on this fact, wavelet coefficients corresponding to the signal will be greater than the noise after wavelet decomposition. The noise can be suppressed by selecting a suitable threshold and properly processing the wavelet coefficients and the main signal features can be preserved. Thus, the key factors in wavelet thresholding are threshold estimation and construction of the thresholding function [bib_ref] A particle swarm optimization technique-based parametric wavelet thresholding function for signal de-noising, Zhang [/bib_ref]. That is, after a suitable threshold is selected, an appropriate thresholding function can be used to compress the wavelet coefficients.
Soft and hard thresholding functions are the most used. Overall discontinuity of hard thresholding functions can lead to abrupt shock points in denoised signals, which is particularly obvious when the noise level is high [bib_ref] A particle swarm optimization technique-based parametric wavelet thresholding function for signal de-noisin, Zhang [/bib_ref]. When a soft thresholding function is used, there will be some deviation between the estimated wavelet coefficient and the real signal wavelet coefficient [bib_ref] A de-noising algorithm to improve SNR of segmented gamma scanner for spectrum..., Li [/bib_ref]. In this paper, a new thresholding function is adopted [bib_ref] A particle swarm optimization technique-based parametric wavelet thresholding function for signal de-noising, Zhang [/bib_ref]. The new thresholding function is a compromise between hard thresholding and soft thresholding. The constant deviation between the estimated wavelet coefficient and the wavelet coefficient of noisy signals can be modified by changing the value of the regulation coefficient α. The thresholding function iŝ
[formula] o j;k ¼ o j;k À sgnðo j;k Þ sin p 2 � l o j;k ! ! 0; ðjo j;k j < lÞ a � l; ðjo j;k j � lÞð1Þ8 > > < > > : [/formula]
where α is the regulation coefficient and λ is the threshold. This new thresholding function has the same continuity as the soft thresholding function in the wavelet domain but approaches the hard thresholding curve as the wavelet coefficients increase [bib_ref] A particle swarm optimization technique-based parametric wavelet thresholding function for signal de-noising, Zhang [/bib_ref]. The wavelet denoising effect is improved by selecting the optimal threshold. The new thresholding function in Eq (1) improves the flexibility of the threshold function and allows the wavelet threshold values to be adaptively selected, to a certain extent. In this paper, the Donoho threshold is used and can be expressed as
[formula] l j ¼ s j ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi 2ln n j qð2Þ [/formula]
where λ j is the wavelet threshold of layer j; n j is the length of the wavelet coefficient on scale j;
[formula] s j ¼ MADðjo j;k j; 0 � k � 2 jÀ 1 Þ=q, MAD(.) [/formula]
is an operator that computes the median value.
The wavelet threshold of each layer can be obtained using the gradient iteration method. Long iteration times and complex characteristics of roll eccentricity signals will reduce the denoising effect. In this paper, the modified particle swarm optimization (MPSO) algorithm is used to search for the optimal threshold and shape adjustment parameters using the root mean square error (RMSE) between the original signal and denoised signal as fitness function.
## Wavelet threshold optimized by mpso
Particle swarm optimization (PSO) has the advantages of a simple concept that is easy to implement and fast convergence. Each particle represents a possible solution to an optimization problem and characteristics of a particle include its fitness value, velocity, and position. The fitness value is calculated using the adaptation function. In each iteration, the particle velocity (V) and position (X) are updated, as follows:
[formula] V kþ1 id ¼ oV k id þ c 1 r 1 ðP k id À X k id Þ þ c 2 r 2 ðP k gd À X k id Þð3ÞX kþ1 id ¼ X k id þ V kþ1 id ð4Þ V id ¼ V max V id > V max V id ¼ À V max V id < À V max ð5Þ ( [/formula]
where ω is the inertia weight; d = 1, 2, . . ., D; i = 1, 2, . . ., n; k is the current number of iterations; c 1 and c 2 are acceleration coefficients; P id is the best position of the individual particle, p gd is the optimal position of the whole particle swarm. The PSO algorithm can easily fall into local optima, resulting in premature convergence. During the solution process, the algorithm considers a variety of information including previous information of the individual, the best position of each particle, and the best position of the total swarm. However, the influence of other individual particle information on particle motion is not considered. Here, a modified particle swarm optimization (MPSO) algorithm is introduced into the threshold determination process [bib_ref] Feng Bin. A global search strategy of quantum behaved particle swarm optimization, Jun [/bib_ref]. The particle velocity in Formula (3) is updated as follows:
[formula] V kþ1 id ¼ oV k id þ c 1 r 1 ðP k nd À X k id Þ þ c 2 r 2 ðP k gd À X k id Þð6Þ [/formula]
where P nd ¼ 1 n X n i¼1 P id is the average best position of all individual particles.
The main advantage of MPSO is the solution process, which seeks the optimal particle solution. Each particle not only obtains its own optimal position information, but also learns from information about other individual particles in the group. In this way, the particle search direction is determined using more effective information, leading to faster convergence rates. The MPSO algorithm is found to have higher search accuracy and stronger optimization ability, with great improvements in stability and convergence speed compared with the traditional PSO algorithm [bib_ref] A Modified Particle Swarm Optimization Based on Wavelet Mutation, Muxin [/bib_ref].
In general, the root mean square error (RMSE) between reconstructed signals and original signals is the standard measure of a reconstructed signal quality, expressed as RMSE ¼ ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi
[formula] 1 N X N i¼1 f ðiÞ Àf ðiÞ h i 2 v u u tð7Þ [/formula]
where f(i) is the original signal andf i ð Þ is the denoised signal.
According to Eq (1), when threshold λ and regulation coefficient α are selected, the wavelet coefficient of the thresholding function can be obtained; that is, the reconstructed signal can be determined after denoising. Therefore, the vector composed of threshold λ and coefficient α can be regarded as the particle position in the MPSO algorithm and represents a potential solution vector. The corresponding fitness value is obtained using Eq (7).
The MPSO process for optimizing the wavelet threshold algorithm is illustrated in [fig_ref] Fig 1: Flow chart of MPSO-WTD algorithm [/fig_ref] The specific steps are as follows:
Step 1: Signal preprocessing. Collect the rolling force signal, select the appropriate wavelet basis function, and determine the number of decomposition layers N. If the number of decomposition layers is too large, threshold processing of the wavelet spatial coefficients of each layer will cause serious signal information losses, the signal-to-noise (SNR) ratio will be reduced, and the computing speed will be slow. Similarly, an insufficient number of decomposition layers will also affect the final denoising effect. Here, the SNR of the denoised signal is repeatedly calculated using Eq (8) and the number of decomposition layers N is optimal. The SNR is calculated as
[formula] SNR ¼ 10log 10 f ðiÞ f ðiÞð8Þ [/formula]
Step 2: Signal decomposition. Decompose the initialized signal to obtain the wavelet coefficients of each layer.
Step 3: Parameter optimization. Optimize the wavelet coefficients of each layer using the MPSO algorithm. The specific method is as follows: Initialize the search space and position of the MPSO algorithm. Different thresholds should be set for different layers, in which the particle position consists of threshold λ and regulation coefficient α.
(1) Calculate the fitness value of each particle according to Eq (7). (2) Use Eqs (4) and (6) to update the velocity and position of the particle.
(3) Update position P nd as the average of the best position of all particles in the total population and optimal value P i of the current position of each particle. Compare the new value with the previous optimal value.
(4) Repeat steps (2)-(4). The optimal value of the wavelet threshold of the current layer j is obtained when the termination condition or maximum number of iterations is reached.
Step 4: Signal reconstruction. Use the optimal threshold λ and regulation coefficient α of each decomposition layer to denoise the signal and output the denoised signal.
# Itd method
The ITD method is self-adaptive and can decompose complex non-stationary signals into several PRCs and a monotonic trend component. The method calculates the instantaneous frequency and amplitude of each rotation component in each local period using a piecewise method, which can overcome mode aliasing and realize real-time data processing [bib_ref] Joint amplitude and frequency demodulation analysis based on intrinsic timescale decomposition for..., Feng [/bib_ref]. Thus, the ITD method is suitable for online signal data processing. In addition, the baseline (mean curve) definition is derived through linear transformation of signals, which can shorten the calculation time and reduce error in the fitting process, thereby achieving high calculation efficiency and high calculation accuracy. The basic steps of ITD are described below, where L is defined as the baseline extraction operator for signal X t .
Defining L t = LX t , the operator can be used to represent the baseline curve of the signal with X t = L t + H t , where H t is defined as a reasonable PRC.
The specific operation process can be described as follows:
(1) Determine the extreme value X k and corresponding time τ k of signal X t (t � 0), where M is the number of extreme points.
(2) Determine extraction operator L of the piecewise linear baseline of the signal:
[formula] L ¼ L k þ ðL kþ1 À L k =L kþ2 À L k ÞðX t À X k Þ; t 2 ðt k ; t kþ1 Þð9Þ [/formula]
with
[formula] L kþ1 ¼ a X k þ t kþ1 À t k t kþ2 À t k � � ðX kþ2 À X k Þ � � þ ð1 À aÞX kþ1 k ¼ 1; 2; . . . ; M À 2 [/formula]
where 0 < α < 1. Normally, α is 0.5.
(3) Define the PRCs used to extract the operator.
[formula] H 1 t ¼ HX t ¼ X t À LX t ¼ X t À L 1 tð10Þ [/formula]
where H 1 t is the PRC component with the highest frequency and baseline signal L 1 t can be used as the initial signal.
## (4) repeat steps (1)-(3) until the baseline signal is a monotone function or a constant function. then the original signal can be decomposed into
[formula] X t ¼ LX t þ HX ¼ HX t þ ðH þ LÞLX t ¼ ½Hð1 þ LÞ þ L 2 �X t ¼ H X PÀ 1 K¼0 L K þ L P " # X t ¼ H 1 t þ H 2 t þ H 3 t þ . . . H P t þ Rð11Þ [/formula]
where H P t is the pth rotation component and R is a monotone function or residual term. In the signal decomposition process with ITD, the local mean value of the signal is calculated by extracting the local extreme value of the signal. However, the distribution of extreme points of the signal will be affected by environmental noise and the anti-noise performance of the ITD method is poor. To improve the accuracy of signal feature extraction, it is necessary to eliminate noise in the ITD signal as much as possible. Therefore, the MPSO-WTD method is adopted as a denoising pretreatment.
## Simulation experiment
The roll eccentricity signal of a cold rolling mill can be represented by the rolling force signal, roll gap signal, and other signals. Therefore, the roll eccentricity signal can be regarded as a series of superimposed high-frequency sinusoidal periodic waves and a complex signal composed of random noise signals. The frequency depends on the speed of the support roller [bib_ref] Roll eccentricity extraction and compensation based on improved wavelet denoising and EEMD, Li-Xin Wei [/bib_ref]. Here, the influence of roller thermal deformation and wear on the amplitude of the eccentricity signal is not considered. The eccentricity signal of a four-high non-reversible cold rolling mill is defined as f ðtÞ ¼ 0:03sinð10t þ 7:3Þ þ 0:018sinð9:23t þ 15:2Þ þ nðtÞ ð12Þ
where n(t) is a random noise signal. In this study, the cofi5 wavelet basis function was used to decompose the signal. The number of decomposition layers was 7 and the signal was decomposed using the above method. The threshold value (λ) and the adjusting parameter (α) were determined using an iterative method and the MPSO algorithm, respectively. The reconstructed signal after denoising is shown in [fig_ref] Fig 2: Comparison of signal denoising effect [/fig_ref] Compared with the gradient iteration-wavelet denoising algorithm, the MPSO-WTD algorithm can better retain the original signal information. The RMSE of the results obtained using the MPSO-WTD algorithm and iterative method were 0.16 and 0.28, respectively, and the corresponding SNRs were 9.13 and 7.24. The results show that MPSO-WTD not only improves the denoising effect of eccentricity signals, but also preserves singularity of the original signal. [fig_ref] Fig 3: Approximated waveforms using coefficients of each layer of the wavelet decomposition [/fig_ref] approximate waveforms obtained using the coefficients of each layer of the wavelet decomposition. It is known that d5 can reflect the roll eccentricity signal by Fourier analysis. The amplitude-frequency characteristics of d5 are shown in [fig_ref] Fig 4: Amplitude-frequency characteristics of d5 [/fig_ref] The signal extracted using the wavelet decomposition method has unrelated frequency information, which will result in interference in the reconstructed eccentricity signal.
The signal model presented in Eq [bib_ref] Online Monitoring System Design for Roll Eccentricity in Rolling Mills, Xu [/bib_ref] was decomposed by the ITD method and the result is shown in The amplitude-frequency characteristics of each PRC are shown in [fig_ref] Fig 6: Amplitude-frequency characteristics of each proper rotation component [/fig_ref] While PRC2 contains eccentricity signals of 1.6 Hz and 1.45 Hz, other spurious frequencies can be observed. Moreover, an eccentricity signal with a frequency of 1.6 Hz is present in PRC1 and PRC3. The results show that noise will disturb the distribution of extreme points in the signal in the ITD process; therefore, the accuracy of the calculated results will be affected. To improve the accuracy of signal feature extraction, it is necessary to eliminate the influence of noise signals in eccentricity signal extraction by ITD as much as possible.
The MPSO-WTD-ITD method was used to extract the roll eccentricity from the model presented in Formula [bib_ref] Online Monitoring System Design for Roll Eccentricity in Rolling Mills, Xu [/bib_ref]. The results are shown in Figs 7 and 8. The eccentricity signal was decomposed into one PRC component and one residual component, as shown in [fig_ref] Fig 7: Signal decomposition by MPSO-WTD-ITD [/fig_ref] Fourier analysis was performed on the PRC component, as shown in [fig_ref] Fig 8: Amplitude-frequency characteristics [/fig_ref] The results are consistent with those obtained by the wavelet and ITD algorithms. The proposed MPSO-WTD-ITD method can suppress spectrum aliasing and spectrum chaos phenomena in wavelet decomposition, reduce the influence of noise signals, and accurately extract the characteristic frequency of the eccentricity signal.
## Experimental verification
## Experimental setup
A four-high cold rolling mill was used in the experiment, as shown in A block diagram of the test control system is presented in [fig_ref] Fig 1: Flow chart of MPSO-WTD algorithm [/fig_ref] Tests were carried out using a pressure sensor, encoder, data acquisition card, and data processor. The data acquisition card was used to collect the rolling force signal measured by the pressure sensor. The processing unit of the eccentricity compensation signal was then used to obtain the roll eccentricity compensation signal. The data processing unit determines the time of sampling and the time interval for sampling according to the number of pulses obtained from the encoder. Finally, the eccentricity compensation signal, position of the hydraulic cylinder obtained by the sensor, and set value of the roll gap were determined. The output signal was amplified by a servo amplifier and sent to the servo valve to control the hydraulic cylinder.
## Test procedure 4.2.1 extraction of roll eccentricity signal.
The easiest way to measure roll eccentricity is by recording fluctuations in the rolling force under rolling mill preloading [bib_ref] Fault Diagnosis of Gearbox based on ITD-Tunable Q-Factor Wavelet Transform, Verma Jay Govind [/bib_ref]. The acquisition period was 0.01 s and the rolling speed was 3.5 m/s. A representative rolling force signal is shown in [fig_ref] Fig 1: Flow chart of MPSO-WTD algorithm [/fig_ref]
## Compensation of roll eccentricity. 4.2.2.1 determination of signal sampling time.
To accurately calibrate the compensation signal to the corresponding position of the roll, the number of sampling points and the number of eccentricity signal compensation points should be the same after one complete rotation cycle of the roll. To prevent changes in rolling speed from affecting the accuracy of eccentricity compensation at each compensation position, an encoder was installed on the support roll. The data processing unit determines the sampling time according to the number of encoder pulses. For example, if the control cycle of the system is 0.01 s, the roll rotates once every 0.6 s; the encoder will emit 1200 pulses, and the system will collect samples at 20-pulses intervals. When the rotation time of the roll changes to 0.5 s, the system will collect samples at 24-pulses interval. The number of sampled points is the number of points compensated within the compensation period of the current eccentricity signal.
## Eccentricity signal conversion.
In this study, the AGC system of the cold rolling mill adopted a closed-loop position control method. The roll gap value was controlled by controlling the hydraulic cylinder displacement. Therefore, it was necessary to convert the extracted rolling force eccentricity signal 4P e into displacement compensation quantity e using the following formula:
[formula] e ¼ K M þ K S K M K S DP eð13Þ [/formula]
where K M is the plasticity coefficient of the rolled piece and K S is the rolling mill stiffness. The roll eccentricity signal compensation process with MPSO-WTD-ITD is shown in [fig_ref] Fig 1: Flow chart of MPSO-WTD algorithm [/fig_ref] The specific steps of the process can be summarized as follows:
(1) The rolling force signal was collected from the rolling mill under the no-load condition.
The rolling thickness, fluctuation in hardness, and other factors were not considered. The signal was preprocessed with MPSO-WTD for denoising.
(2) The ITD method was used to extract the rolling force eccentricity signal.
(3) According to Eq (13), the extracted rolling force eccentricity signal was converted into the displacement compensation signal.
(4) The compensation signal was input into the AGC system of the rolling mill, and a pointto-point eccentricity signal of the roll was realized according to the number of signal pulses output by the encoder.
(5) The output displacement was obtained according to the compensation signal, set value of the roll gap, and actual position of the hydraulic cylinder, and the position of the hydraulic cylinder of the hydraulic press was controlled by the servo amplifier and electro-hydraulic servo valve.
As shown in [fig_ref] Fig 1: Flow chart of MPSO-WTD algorithm [/fig_ref] the roll eccentricity compensation signal was obtained using the wavelet algorithm and MPSO algorithm, and can be divided into 56 equally spaced points.
## Analysis of test results
The controlled plant in the experiment was a four-high cold rolling mill. The work roll diameter was 300 mm, and the backup roll diameter was 600 mm. The rolling speed was 3.5 m/s. The entrance thickness of the aluminum alloy 3004 strip was 1 mm, and the outlet thickness set-point was 0.52 mm. During the experiment, the rolling force data were updated with the newly sampled data in each control step. the rolling speed is not uniform due to high precision of the roll eccentricity signal extraction process and control system.
The roll eccentricity compensation effect can be expressed as the rate of improvement in strip thickness characteristic R using the following formula:
[formula] R ¼ H e À H Hð14Þ [/formula]
where H e is range of strip thickness fluctuation when roll eccentricity is controlled and H is the range of strip thickness fluctuation when roll eccentricity is not controlled. The eccentricity compensation achieved by the two methods was input into the automatic gauge control (AGC) system of the cold rolling mill. The effects of roll eccentricity compensation in the stable rolling stage are presented in [fig_ref] Table 1: Effects of roll eccentricity compensation in stable rolling stage [/fig_ref]. A total of 100 strip thickness values were collected for normal distribution statistical analysis. It can be concluded that the probability of a strip thickness error of ±3.5μm is about 96%, and the roll eccentricity compensation effect reaches 62.3% during steady-state rolling.
# Conclusions
A roll eccentricity extraction method based on MPSO-TWD and ITD was proposed to improve the accuracy of roll eccentricity signal extraction. Simulations were carried out the roll eccentricity compensation signal and were also input into the AGC system of a four-high cold rolling mill to verify the results. The main conclusions of this study can be summarized as follows:
(1) The gradient iteration method and MPSO algorithm were used to calculate the threshold and coefficient of wavelet, respectively. The simulation results show that the MPSO-WTD method has better denoising effects. Comparing the denoising effects of ITD, MPSO-WTD, and MPSO-WTD-ITD on roller eccentricity signals, the simulation results show that the proposed method avoids the frequency aliasing phenomenon of wavelet analysis and poor anti-noise performance of the ITD method. The proposed method has high precision in extracting roller eccentricity signals.
(2) To avoid the influence of rolling speed on the frequency of the roll eccentricity signal, the frequency of the eccentricity compensation signal is determined according to the number of encoder pulses. Due to the complexity of the rolling process and accuracy limitations of the test equipment, the strip thickness control will be affected. However, the experimental results show that the proposed method is satisfactory. It can be concluded that 96% of the strip thickness errors can be controlled to ±3.5μm and the roll eccentricity compensation effect reaches 62.3% during steady-state rolling by collected 100 strip thickness. Supporting information S1 Data.
(XLSX)
[fig] Fig 1: Flow chart of MPSO-WTD algorithm. https://doi.org/10.1371/journal.pone.0259810.g001 [/fig]
[fig] Fig 2: Comparison of signal denoising effect. https://doi.org/10.1371/journal.pone.0259810.g002 [/fig]
[fig] Fig 3: Approximated waveforms using coefficients of each layer of the wavelet decomposition. https://doi.org/10.1371/journal.pone.0259810.g003 [/fig]
[fig] Fig 4: Amplitude-frequency characteristics of d5. https://doi.org/10.1371/journal.pone.0259810.g004Fig 5. Intrinsic time-scale decomposition (ITD). https://doi.org/10.1371/journal.pone.0259810.g005 [/fig]
[fig] Fig 6: Amplitude-frequency characteristics of each proper rotation component (PRC).https://doi.org/10.1371/journal.pone.0259810.g006 [/fig]
[fig] Fig 7: Signal decomposition by MPSO-WTD-ITD. https://doi.org/10.1371/journal.pone.0259810.g007 [/fig]
[fig] Fig 8: Amplitude-frequency characteristics. https://doi.org/10.1371/journal.pone.0259810.g008Fig 9. Four-high cold rolling mill. https://doi.org/10.1371/journal.pone.0259810.g009 [/fig]
[table] PLOS ONE: | https://doi.org/10.1371/journal.pone. [/table]
[table] Table 1: Effects of roll eccentricity compensation in stable rolling stage. https://doi.org/10.1371/journal.pone.0259810.t001 [/table]
|
Effects of tourniquet inflation on peri- and postoperative cefuroxime concentrations in bone and tissue
Background and purpose -Tourniquet is widely used in orthopedic surgery to reduce intraoperative bleeding and improve visualization. We evaluated the effect of tourniquet application on peri-and postoperative cefuroxime concentrations in subcutaneous tissue, skeletal muscle, calcaneal cancellous bone, and plasma. The primary endpoint was the time for which the free cefuroxime concentration was maintained above the clinical breakpoint minimal inhibitory concentration (T > MIC) for Staphylococcus aureus (4 µg/mL).Patients and methods -10 patients scheduled for hallux valgus or hallux rigidus surgery were included. Microdialysis catheters were placed for sampling of cefuroxime concentrations bilaterally in subcutaneous tissue, skeletal muscle, and calcaneal cancellous bone. A tourniquet was applied on the thigh of the leg scheduled for surgery (tourniquet duration time [range]: 65 minutes [58-77]). Cefuroxime (1.5 g) was administered intravenously 15 minutes prior to tourniquet inflation, followed by a second dose 6 hours later. Dialysates and venous blood samples were collected for 12 hours.Results -A cefuroxime concentration of 4 µg/mL was reached within 23 minutes in all compartments and patients. For cefuroxime the T > MIC (4 µg/mL) ranged between 4.8 and 5.4 hours across compartments, with similar results for the tourniquet and non-tourniquet leg. Comparable T > MIC and penetration ratios were found for the first and second dosing intervals.Interpretation -Administration of cefuroxime (1.5 g) 15 minutes prior to tourniquet inflation is safe in order to achieve tissue concentrations above 4 µg/mL throughout surgery. A tourniquet application time of approximately 1 hour did not affect the cefuroxime tissue penetration in the following dosing interval.
## Effects of tourniquet inflation on peri-and postoperative cefuroxime concentrations in bone and tissue
Background and purpose -Tourniquet is widely used in orthopedic surgery to reduce intraoperative bleeding and improve visualization. We evaluated the effect of tourniquet application on peri-and postoperative cefuroxime concentrations in subcutaneous tissue, skeletal muscle, calcaneal cancellous bone, and plasma. The primary endpoint was the time for which the free cefuroxime concentration was maintained above the clinical breakpoint minimal inhibitory concentration (T > MIC) for Staphylococcus aureus (4 µg/mL).
Patients and methods -10 patients scheduled for hallux valgus or hallux rigidus surgery were included. Microdialysis catheters were placed for sampling of cefuroxime concentrations bilaterally in subcutaneous tissue, skeletal muscle, and calcaneal cancellous bone. A tourniquet was applied on the thigh of the leg scheduled for surgery (tourniquet duration time : 65 minutes . Cefuroxime (1.5 g) was administered intravenously 15 minutes prior to tourniquet inflation, followed by a second dose 6 hours later. Dialysates and venous blood samples were collected for 12 hours.
Results -A cefuroxime concentration of 4 µg/mL was reached within 23 minutes in all compartments and patients. For cefuroxime the T > MIC (4 µg/mL) ranged between 4.8 and 5.4 hours across compartments, with similar results for the tourniquet and non-tourniquet leg. Comparable T > MIC and penetration ratios were found for the first and second dosing intervals.
Interpretation -Administration of cefuroxime (1.5 g) 15 minutes prior to tourniquet inflation is safe in order to achieve tissue concentrations above 4 µg/mL throughout surgery. A tourniquet application time of approximately 1 hour did not affect the cefuroxime tissue penetration in the following dosing interval. Tourniquet (TQ) is widely used in orthopedic surgery due to its ability to reduce intraoperative bleeding and improve visualization. However, as the blood supply to the operating field is occluded during surgery, correct timing of antimicrobial prophylaxis administration and TQ inflation is essential in order to ensure therapeutic tissue concentrations at the site of surgery. Only a few studies have investigated the ideal time interval from perioperative antimicrobial prophylaxis administration to TQ inflation, resulting in ambiguous guidelines. With regard to cefuroxime in particular, a recent randomized controlled microdialysis study in a pig model suggested that a window of 15-45 minutes between cefuroxime administration and TQ inflation results in sufficient perioperative tissue concentrations throughout a 90-minute TQ application.
TQ induces peri-and postoperative ischemia, which may result in decreased postoperative tissue perfusion and antimicrobial tissue concentration. A recent study on a rat model demonstrated a reduced distribution of antimicrobials to TQ-affected tissues for up to 72 hours after TQ release. Decreased postoperative antimicrobial tissue concentrations may ultimately increase the risk of surgical site infection. Therefore, we dynamically evaluated effects of TQ application on both peri-and postoperative in situ cefuroxime concentrations in subcutaneous tissue, skeletal muscle, calcaneal cancellous bone, and plasma. Cefuroxime (1.5 g) was administered intravenously prior to TQ inflation and followed by a subsequent dose 6 hours later. The primary aim was to assess the time for which the free drug concentration of cefuroxime was maintained above the clinical breakpoint minimal inhibitory concentration (T > MIC) for Staphylococcus aureus (4 µg/mL) (EUCAST 2021), which we hypothesized was maintained throughout surgery in the TQ-exposed tissues when administering cefuroxime 15 minutes prior to tourniquet inflation.
# Patients and methods
This study was conducted at the Department of Orthopedic Surgery, Horsens Regional Hospital, Denmark. Chemical analyses were performed at the Department of Clinical Biochemistry, Aarhus University Hospital, Denmark. This study was performed in the same setting as another study, which investigated tissue ischemic metabolites.
## Study procedure microdialysis
The microdialysis catheter consists of a semipermeable membrane at the tip of the catheter, which allows for sampling of water-soluble molecules such as antimicrobials. However, as the semipermeable membrane is continuously perfused, equilibrium across the semipermeable membrane cannot be attained. Consequently, the dialysates represent only a fraction of the actual tissue concentration. This fraction is referred to as the relative recovery that can be determined by different calibration methods. For this study, meropenem was used as an internal calibrator for cefuroxime. An in-depth description of the micro-dialysis technique and the equation for calculating the relative recovery can be found elsewhere.
We used microdialysis equipment from M Dialysis AB (Stockholm, Sweden). The microdialysis catheters consisted of CMA 63 membranes and CMA 107 precision pumps (flow rate: 2 µL/min).
## Study design and patients
10 patients were included in this prospective observational cohort study. The effects of TQ application on both peri-and postoperative cefuroxime concentrations were evaluated in subcutaneous tissue, skeletal muscle, and calcaneal cancellous bone in a simultaneous paired comparison of the TQ and non-TQ leg during 12 hours of continuous microdialysis sampling .
Patients scheduled for hallux valgus or hallux rigidus surgery were offered enrolment in the study. A single surgeon recruited 10 patients who attended the outpatient clinic. Inclusion and exclusion criteria are presented in. All patients invited for enrolment were included in the study and all completed the study.
After placement of the 6 microdialysis catheters, cefuroxime (1.5 g) (FreseniusKabi AB, Sweden) was administered intravenously over 10 minutes, marking time 0. 15 minutes after initiation of the cefuroxime administration, the TQ cuff was inflated (pressure 260 mmHg) on the thigh of the leg scheduled for surgery. Prior to TQ inflation, the leg was elevated for 1 minute. The planned surgical procedure was performed after TQ inflation. When the surgical procedure was completed, the TQ cuff was released (mean TQ inflation time [range]: 65 minutes [58-77]). A second dose of 1.5 g cefuroxime was administered at 6 hours.
## Surgery
Before the surgical procedure, microdialysis catheters were placed similarly in both legs: in the subcutaneous tissue (membrane length 30 mm), at the posterior site of the mid-lower leg, in the gastrocnemius muscle of the medial head (membrane length 30 mm), and in the calcaneal cancellous bone (mem- . Illustration of the inserted microdialysis catheters. Cefuroxime concentrations were obtained by means of microdialysis catheters placed in non-tourniquet subcutaneous tissue (1), non-tourniquet skeletal muscle (2), non-tourniquet calcaneal cancellous bone (3), tourniquet subcutaneous tissue (4), tourniquet skeletal muscle (5), and tourniquet calcaneal cancellous bone (6). A tourniquet cuff (7) was placed on the leg scheduled for surgery. brane length 10 mm) via drill holes (ø: 2 mm; depth 30 mm) made on the posterolateral side, aiming at the anteromedial side of the calcaneal bone . After placement of the microdialysis catheters, all catheters were perfused with 0.9% NaCl containing 5 µg/mL meropenem, allowing for continuous calibration with meropenem as an internal calibrator.
## Sampling procedures
Dialysates were collected from all 6 microdialysis catheters at 15-minute intervals from time 0-30 minutes, at 30-minute intervals from time 30-180 minutes, and at 60-minute intervals from both time 180-240 minutes and time 300-360 minutes. Following administration of the second dose of 1.5 g cefuroxime at time 360 minutes, dialysates were collected at 30-minute intervals from time 360-540 minutes, and at 60-minute intervals from both time 540-600 minutes and time 660-720 minutes. 17 samples from each microdialysis catheter were collected over the 12-hour period. Venous blood samples were collected at the midpoint of the sampling intervals drawn from a peripheral catheter in the cubital vein. After the last sample was collected, all microdialysis catheters were removed.
## Handling of samples
The venous blood samples were stored at 5°C for a maximum of 10 hours before being centrifuged at 3,000 g for 10 minutes. The plasma aliquots were then stored at -80°C until analysis. The dialysate samples were immediately stored at -80°C until analysis.
## Quantification of cefuroxime and meropenem concentrations
The concentrations of cefuroxime and meropenem were quantified using a validated ultra-high-performance liquid chromatography assay . Inter-run imprecisions (% coefficients of variation) were 4.7% at 2.5 µg/mL for quantification of cefuroxime and 3.0% at 2.0 µg/mL for quantification of meropenem. The lower limits of quantification were 0.06 µg/mL for cefuroxime and 0.5 µg/mL for meropenem.
## Pharmacokinetic analysis and statistics
The cefuroxime concentrations of the dialysate were attributed to the midpoint of the sampling intervals. Pharmacokinetic parameters, areas under the concentration-time curves (AUC), peak drug concentration (C max ), time to C max (T max ), and T 1/2 , were determined separately for each compartment for each patient by non-compartmental analysis using the pharmacokinetic series of commands in Stata (v. 15.1, Stata-Corp, College Station, TX, USA). AUC were calculated using the linear up-log down trapezoidal method. The maximum of all the recorded concentrations was defined as C max , enabling calculation of T max . The T 1/2 was calculated as ln(2)/λ eq , where λ eq is the terminal elimination rate constant estimated by linear regression of the log concentration on time. The AUC tissue / AUC plasma ratio was calculated as a measure of tissue penetration. Microsoft was used to estimate the T > MIC (4 µg/mL) using linear interpolation. The pharmacokinetic parameters and T > MIC were calculated separately for both the first (time 0-6 hours) and second (time 6-12 hours) dosing intervals. The pharmacokinetic parameters and T > MIC were obtained in all 7 compartments from the same patient and a mixed model for repeated measurements had compartments as fixed effect and subject identification variable as a random effect was applied. Also, distinct residual variance was assumed within each compartment. The normality of the residuals was estimated using a quantile-quantile plot for the residuals and the homogeneity of the residual variance was checked by plotting residuals vs. best linear unbiased prediction estimates. The normality of the estimated random effects was checked using a quantile-quantile plot of the estimated random effects. The Kenward-Roger approximation method was used for degrees of freedom correction due to the small sample size. The F-test was used to determine the overall comparisons between the compartments and a t-test was used to determine pairwise comparisons. A significance level of 5% was used. Statistical analyses were performed using Stata. Foundation, and the Familien Hede Nielsen Foundation. The funding sources did not have any roles in the investigation, data interpretation, or paper presentation. The authors have no conflicts of interest.
## Sample size
# Results
The patients' characteristics are presented in. No adverse events related to the microdialysis technique or cefuroxime infusion occurred.
The mean relative recovery (SD) values were 23% (9) for TQ subcutaneous tissue, 20% (7) for non-TQ subcutaneous tissue, 39% (4) for TQ skeletal muscle, 33% (12) for non-TQ skeletal muscle, 21% (8) for TQ calcaneal cancellous bone, and 19% (7) for non-TQ calcaneal cancellous bone.
## T > mic
Comparable results were observed for T > MIC (4 µg/mL) between the first and second dosing intervals. Therefore, the T > MIC results are presented only for the first dosing interval in. A cefuroxime concentration of 4 µg/mL was reached within 23 minutes in all compartments and patients. The T > MIC (4 µg/mL) ranged between 4.8 and 5.4 hours across compartments, and no significant differences were found between the TQ and non-TQ exposed legand. When comparing TQ and non-TQ legs separately, lower T > MIC values were found for calcaneal cancellous bone compared with the remaining compartments in the TQ leg, including plasma (p < 0.05). No differences were found between the compartments in the non-TQ leg.
## Pharmacokinetic parameters
Comparable pharmacokinetic results were seen between the first and second dosing intervals in all investigated compartments. Only the TQ calcaneal cancellous bone T max was longer in the first dosing interval (mean No statically significant differences were observed for AUC, C max , T 1/2 , and tissue penetrations when comparing the TQ and non-TQ leg. Only the calcaneal cancellous bone T max was longer in the TQ leg (mean [range], 84 minutes [23-135]) compared with the non-TQ leg (mean [range], 35 minutes [22-75]) (p < 0.01) in the first dosing interval. No statistically significant differences were found for the remaining compartments.
When comparing the TQ and non-TQ leg separately, a lower AUC was found for the non-TQ calcaneal cancellous bone compared with non-TQ subcutaneous tissue. Plasma C max was higher compared with all investigated compartments. Moreover, the TQ skeletal muscle C max was higher compared with both TQ calcaneal cancellous bone and TQ subcutaneous tissue. Finally, plasma T max was shorter compared with all tissues in both the TQ and non-TQ leg, and the TQ calcaneal cancellous bone was longer than both TQ subcutaneous tissue and TQ skeletal muscle.
# Discussion
This is the first clinical study to investigate the effects of TQ application on both peri-and postoperative cefuroxime concentrations in subcutaneous tissue, skeletal muscle, and calcaneal cancellous bone in a simultaneous paired comparison of the TQ and non-TQ leg. Our main finding was a cefuroxime T > MIC (4 µg/mL) range between 4.8 and 5.4 hours across compartments. Furthermore, comparable T > MIC and penetration ratios were found for the first and second dosing intervals.
TQ is widely used in orthopedic surgery, but only a few studies have investigated antimicrobial tissue concentrations during the TQ application, and no clinical studies have investigated antimicrobial tissue concentrations after TQ release. Using bone and fat tissue specimens,investigated different time intervals from administration of cefuroxime (1.5 g) to TQ inflation, and concluded that a time interval of 10 minutes was sufficient to achieve tissue concentrations above 4 µg/mL. Recently, a randomized controlled microdialysis study in a pig model suggested that a window of 15-45 minutes between cefuroxime (1.5 g) administration and TQ inflation was sufficient to achieve calcaneal cancellous bone and subcutaneous tissue concentrations above 4 µg/mL. The present clinical study confirms these findings, suggesting that cefuroxime has fully penetrated the investigated tissues after 15 minutes.
It has previously been hypothesized that perioperative ischemia reduces the postoperative antimicrobial tissue penetration . However, studies investigating tissue ischemia during and after TQ application found that ischemia-exposed tissue fully recovers 2.5 hours after TQ release. Our findings do not indicate any decreased postoperative cefuroxime penetration in the TQ exposed tissues for a TQ application of approximately 1 hour.
Interestingly, our study showed that TQ calcaneal cancellous bone T max is longer than in non-TQ calcaneal cancellous bone. Furthermore, a wider range of the T max values was found for both subcutaneous tissue and skeletal muscle in the TQ leg compared with the non-TQ leg. These T max results may be attributed to a combination of the limited elimination rate of cefuroxime during TQ time and a second peak in the cefuroxime concentration after TQ release. For 5 patients, this peak was higher than the initial peak prior to TQ inflation in TQ calcaneal cancellous bone. This may indicate a favorable hyperemic effect when the TQ is released, which was also observed in a pig model.
For antimicrobial prophylaxis it is generally recommended that the antimicrobial plasma and tissue concentrations exceed the MIC values of relevant bacteria throughout surgery. In our study a TQ cuff was inflated 15 minutes after initiation of the cefuroxime admin- . Pharmacokinetic parameters for plasma, subcutaneous tissue, skeletal muscle, and calcaneal cancellous bone on both the tourniquet and non-tourniquet leg
## Compartment
Non-tourniquet Tourniquet p-value AUC 0-6h , mean (95% CI), 10 4 min μg/mL Plasma 8.2 (6.6-9.8) --Subcutaneous tissue 8.5 (7.0-10) 7.6 (6.0-9.1) 0.3 Skeletal muscle 7.3 (5.7-8.9) 7.8 (6.2-9.4) 0.6 Calcaneal canc. bone 6.6 (5. istration and a cefuroxime concentration of 4 µg/mL was reached within 23 minutes in all tissues and patients, which was maintained above this target for a minimum of 4.5 hours in all the investigated compartments. As such, these findings indicate that cefuroxime appears to be a good choice for antimicrobial prophylaxis in terms of tissue penetration and T > MIC. Only 1 clinical study has previously investigated cefuroxime bone tissue concentrations by means of microdialysis.found a shorter T > MIC in plasma, subcutaneous tissue, and tibial cancellous bone after a postoperative intravenous bolus administration of 1.5 g cefuroxime compared with our study compartments. While the plasma creatinine was comparable between the patient groups in the 2 studies, Tottrup et al. recorded a substantially higher mean BMI compared with the present study (31 vs. 25). Weight-based dosing of cefuroxime, in addition to consideration of renal function, may therefore be considered in order to achieve therapeutic tissue concentrations in heavy patients.
The few clinical studies that have investigated antimicrobial concentrations during TQ application have been based on tissue specimens. However, this approach suffers from important methodological limitations because sampling in clinical studies is limited to the time of surgery, free extracellular concentrations cannot be measured selectively, and drug concentrations are given by mass rather than volume. Microdialysis, on the other hand, allows for simultaneous and serial sampling of the free and active fraction of drugs in the interstitial space from multiple compartments, both peri-and postoperatively. These features are desirable, as the majority of infections occur in the interstitial space. However, microdialysis remains a sampling technique that has limitations associated with calibration procedures and chemical assays. The major limitation of our study is the small sample size. Although a paired design and no statistically significant differences for the T > MIC (4 µg/mL) between TQ and non-TQ exposed tissues were demonstrated for this specific patient population, a larger study population may alter these findings. However, as all mean tissue cefuroxime concentrations were above 4 µg/mL approximately 4-5 times longer than the presented surgery time, any potential difference between the TQ and non-TQ exposed tissue may be without clinical relevance.
In summary, administering cefuroxime (1.5 g) 15 minutes prior to TQ inflation seems safe in order to achieve tissue concentrations above 4 µg/mL throughout 1 hour surgery, as T > MIC (4 µg/mL) ranged between 4.8 and 5.4 hours in subcutaneous tissue, skeletal muscle, and calcaneal cancellous bone. A TQ application time of approximately 1 hour did not affect the cefuroxime tissue penetration in the following dosing interval. PH, MB, JK, KS, and MS initiated and designed the study. PH, JK, and CJ conducted the surgery and placed all the probes. PH, MB and ARJ collected the data. Statistical analysis and interpretation of data was done by PH, MB, JK, KS, and MS. All authors drafted and revised the manuscript. |
Recent advances in understanding and managing resistant/refractory hypertension
F1000 Faculty Reviews are written by members of the prestigious . They are F1000 Faculty commissioned and are peer reviewed before publication to ensure that the final, published version is comprehensive and accessible. The reviewers who approved the final version are listed with their names and affiliations.AbstractThe management of resistant hypertension presents several challenges in everyday clinical practice. During the past few years, several studies have been performed to identify efficient and safe pharmacological and non-pharmacological options for the management of such patients. The Spironolactone versus placebo, bisoprolol, and doxazosin to determine the optimal treatment for drug-resistant hypertension (PATHWAY-2) trial demonstrated significant benefits with the use of spinorolactone as a fourth-line drug for the treatment of resistant hypertension over doxazosin and bisoprolol. In addition, recent data support that spironolactone may demonstrate superiority over central acting drugs in such patients, as well. Based on the European guidelines, spironolactone is recommended as the fourth-line drug option, followed by amiloride, other diuretics, doxazosin, bisoprolol or clonidine. Among several device-based approaches, renal sympathetic denervation had fallen into hibernation after the disappointing results of the Renal Denervation in Patients With Uncontrolled Hypertension (SYMPLICITY HTN) 3 trial. However, the technique re-emerged at the epicenter of the clinical and research interest after the favorable results of three sham-controlled studies, which facilitated novel catheters and techniques to perform the denervation. Significant results of iliac anastomosis on blood pressure levels have also been demonstrated. Nevertheless, the technique-related adverse events resulted in withdrawal of this interventional approach. Last, the sympatholytic properties of the carotid baroreceptor activation therapy were associated with significant blood pressure reductions in patients with resistant hypertension, which need to be verified in larger controlled trials. Currently device-based approaches are recommended only in the setting of clinical trials until more safety and efficacy data become available.
# Introduction
The prevalence of resistant hypertension ranges from 5 to 30% on the basis of the definition used by relevant studies 1 . However, the true prevalence of resistant hypertension after applying a strict definition and having excluded causes of pseudo-resistant hypertension is less than 10% of the patients with treated hypertension 1 . Importantly, resistant hypertension is related with higher risk for cardiovascular morbidity and mortality, chronic kidney disease, and other hypertension-mediated target organ damage 2 .
Based on the European guidelines, resistant hypertension is defined as the failure to reduce systolic or diastolic blood pressure (BP) levels (or both) below 140 and 90 mm Hg, respectively, despite treatment with optimal doses (or best-tolerated doses) of an appropriate therapeutic strategy with the triple combination of an angiotensin-converting enzyme inhibitor or an angiotensin receptor blocker with a calcium channel blocker and a thiazide/ thiazide-type diuretic. As stated in the guidelines, home or ambulatory BP measurements should be used to confirm inadequate BP control, and exclusion of pseudo-resistant hypertension and secondary hypertension is mandatory to establish the diagnosis 1 .
In the previous European guidelines (2013), the use of mineralocorticoid receptor antagonists, amiloride, and the alpha-1 blocker doxazosin were considered for the management of resistant hypertension 3 . During the past few years, several studies in resistant hypertension, and especially the spironolactone versus placebo, bisoprolol, and doxazosin to determine the optimal treatment for drug-resistant hypertension (PATHWAY-2) trial 4 , resulted in changes in the recommendations for the management of resistant hypertension. Specifically, the addition of "low-dose spironolactone to existing treatment, or the addition of further diuretic therapy if intolerant to spironolactone, with either eplerenone, amiloride, higher-dose thiazide/thiazide-like diuretic or a loop diuretic, or the addition of bisoprolol or doxazosin" is now recommended 1 . Similarly, the consensus document of the American Heart Association for the management of resistant hypertension recommends the use of either spironolactone or eplerenone as a fourth-line agent, followed by a beta-blocker, a dual beta-and alpha-blocker, clonidine, or diltiazem 5 . The purposes of this review are to report and critically discuss the findings of recent studies that resulted in a stronger recommendation for the treatment of resistant hypertension and to report treatments under investigation that could prove to be useful in such patients.
## Exclusion of pseudo-resistant hypertension
The exclusion of pseudo-hypertension is of paramount importance for the establishment of an accurate diagnosis. BP is often measured inaccurately; wrong-sized cuffs, measurement of BP only once, placement the cuff over the patient's clothes, and wrong position of the patient are common mistakes performed in everyday clinical practice 1 . A 2016 study demonstrated that readings performed in a routine triage setting were higher than the readings performed by trained physicians and resulted in a misdiagnosis of uncontrolled resistant hypertension in 33% of the patients 6 . Under-treatment is also a common cause of pseudoresistant hypertension, and studies indicate that a lack of BP control is often attributable to the absence of treatment intensification 7 .
Another important cause of pseudo-resistant hypertension is poor medication adherence. The recent Renal Sympathetic Denervation as a New Treatment for Therapy Resistant Hypertension (SYMPATHY) trial examined drug adherence with the detection of drug concentrations in blood samples in patients with uncontrolled hypertension on three or more anti-hypertensive drugs or with documented intolerance to two or more of the four major anti-hypertensive drug classes; 16% were non-adherent and 52% were poorly adherent 8 .
## Exclusion of other causes contributing to resistant hypertension
Lifestyle factors such as excessive alcohol and salt intake contribute to the presence of resistant hypertension. Large amounts of alcohol consumption (three or more drinks per day) have an important dose-related effect on BP levels in both normotensive and hypertensive patients 9 . Abstinence in heavy alcohol drinkers may decrease 24 hours systolic and diastolic ambulatory BP levels by up to 7.2 and 6.6 mm Hg, respectively 10 . Usually, patients with resistant hypertension present an average sodium intake that exceeds 10 g per day 11 . Salt not only increases BP levels but also blunts the anti-hypertensive effect of the BP-lowering drugs 12 . In salt-sensitive patients (elders, African-Americans, and patients with chronic kidney disease), these effects are much more pronounced 13 . Moreover, obesity and increased body mass index in general increase significantly BP levels 1 . The mechanisms that induce hypertension in those patients include the activation of sympathetic nervous system and renin-angiotensinaldosterone system and also insulin resistance and impaired sodium excretion 14 . The adoption of current European Society of Cardiology/European Society of Hypertension guidelines on lifestyle changes may significantly decrease BP levels in those patients and contribute to BP control.
Finally, several drugs and substances may increase BP levels. Non-steroidal anti-inflammatory drugs (NSAIDs) represent probably the most common agents in terms of worsening BP control 1 . The use of NSAIDs not only increases BP levels but also can blunt the effect of various anti-hypertensive drugs such as diuretics, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers 1 . The hypertensive effect of NSAIDs is more pronounced in patients with chronic kidney disease 1 . Other substances that can increase BP levels are decongestants and stimulant agents used for weight loss and also contraceptives, cyclosporine, erythropoietin, and cortisone that increase BP levels mainly through fluid retention. A proposed work-up for patients with resistant hypertension is shown in.
## Recent advances in pharmacological therapy
Data from clinical studies of spironolactone versus adrenergic blockers The landmark PATHWAY-2 study found significant benefits with the use of spironolactone in patients with resistant hypertension on a standard three-drug therapy with an angiotensinconverting enzyme inhibitor or an angiotensin receptor blocker, amlodipine, and indapamide 4 . The study was a double-blind four-way crossover study that assessed the use of spironolactone (25 or 50 mg) versus bisoprolol (5 to 10 mg), doxazosin (5 to 10 mg), or placebo. To exclude non-adherence, the study was monitored with pill count and measurement of serum angiotensin-converting enzyme activity. Spironolactone was superior to both active treatments and placebo, and mean reductions in BP were 8.70, 4.48, and 4.03 mm Hg with spironolactone, bisoprolol, and doxazosin, respectively. Importantly, about 60% of the spironolactone users achieved BP control versus 43.3% of the bisoprolol and 41.5% of the doxazosin users 4 .
Apart from the beneficial effect of spironolactone in patients with resistant hypertension, the PATHWAY-2 study offered three important findings. First, despite the superiority of spironolactone over bisoprolol and doxazosin, the use of the latter two drugs was associated with significant reductions in BP compared with placebo. Thus, the European guidelines recommend the use of bisoprolol and doxazosin for the treatment of resistant hypertension when spironolactone is contraindicated or not tolerated 1 . Second, uptitration of spironolactone dose from 25 to 50 mg resulted in a greater reduction in BP at week 12 of the study. The BP-lowering effect of spironolactone uptitration was higher compared with the corresponding increases in the dose of either bisoprolol or doxazosin (−3.86 mm Hg with spironolactone versus −0.88 mm Hg for doxazosin, −1.49 mm Hg for bisoprolol, and −0.68 mm Hg for placebo). Last, while spironolactone reduced BP levels irrespective of renin levels, an enhanced benefit in patients with suppressed renin levels was observed, and there was up to a 20 mm Hg reduction in home BP in patients with the lowest renin levels 4 . Important clinical information for the management of resistant hypertension arose from three recent substudies of the PATH-WAY-2 trial. In the first substudy, the plasma aldosterone, renin, and aldosterone-to-renin ratio were assessed as predictors of home systolic BP response in 126 patients. Plasma aldosterone-to-renin ratio and plasma renin levels were found to be predictors of BP response to spironolactone. In the second one, the impact of each drug on the thoracic fluid content (an index of fluid retention) and vascular resistance was examined (226 patients). Thoracic fluid content was significantly reduced (by 6.8%) from baseline with spironolactone but not with the other active treatments. Given the overall outcomes of the PATHWAY-2 study, this finding supports the theory that patients with resistant hypertension are characterized by volume overload, secondary to aldosterone excess, explaining the greater benefits observed with spironolactone 1 . In the third substudy, the effect of amiloride on systolic BP was examined in a 6-to 12-week open-label runout phase, in which patients on spironolactone were crossed over from spironolactone to amiloride (146 patients). Amiloride use resulted in a remarkable reduction in BP levels of 20.4 mm Hg, comparable to the 18.3 mm Hg observed with spironolactone, suggesting that amiloride might be an effective alternative agent for these patients. Based on these findings, the European guidelines propose the use of amiloride as an alternative option if spironolactone is contraindicated or not well tolerated 15 .
Data from clinical studies of spironolactone versus central acting drugs Although concrete evidence supports the superiority of mineralocorticoid receptor antagonists over alpha-and beta-blockers for patients with resistant hypertension, there is a lack of evidence regarding the use of central acting drugs in such patients. In this setting, the recent Resistant Hypertension Optimal Treatment (ReHOT) study compared the impact of spironolactone and clonidine in 187 patients with resistant hypertension. BP control assessed with office and 24-hour ambulatory BP monitoring was similar across the two groups of patients. However, the 24-hour systolic and diastolic BP reduction and the daytime diastolic BP reductions observed with spironolactone were significantly greater than those observed with clonidine. Given the easier dosage scheme of spironolactone and the greater benefits in various ambulatory BP parameters, spironolactone seems to be a preferable option over clonidine 17 .
Meta-analytic data of mineralocorticoid antagonists versus other drug classes Important information emerged from meta-analytic data for the use of mineralocorticoid receptor antagonists in patients with resistant hypertension. In 869 patients from four trials, spironolactone as add-on therapy was associated with a reduction in BP of 16.67/6.11 mm Hg 16 . A meta-analysis of 662 patients and five trials found that the addition of spironolactone in patients with resistant hypertension resulted in a reduction in office BP levels of 15.73/6.21 mm Hg compared with placebo, but compared with other drugs (beta-blocker, candesartan, or alpha methyldopa), spironolactone reduced home systolic BP by 4.5 mm Hg 18 . In a meta-analysis of five studies and 553 patients with resistant hypertension, spironolactone reduced 24-hour, daytime, nighttime, and office BP by 10.50/4.09, 10.20/4.14, 10.02/3.21, and 16.99/6.18 mm Hg, respectively 19 . Last, a meta-analysis of five studies and 755 patients with resistant hypertension found a greater reduction in systolic BP levels of 7.4 mm Hg (in the randomized studies) and 11.9 mm Hg (in the nonrandomized studies) with mineralocorticoid receptor antagonists compared with other fourth-line options (bisoprolol, doxazosin, furosemide, or other renin-angiotensin system blockers) 20 .
Collectively, accumulating evidence suggests that mineralocorticoid receptor antagonists are the optimal choice as the fourth-line option in patients with resistant hypertension, and data favor their use over central acting drugs, alpha-and betablockers. Most data come from studies with spironolactone. However, eplerenone may be considered as an alternative when adverse effects (such as gynecomastia or vaginal bleeding) are observed with spironolactone therapy 21 , although strong data with eplerenone use are currently missing. Given the substantial fluid retention observed in these patients and the findings of the PATHWAY-2 study, amiloride is an alternative option, while treatment with doxazosin, bisoprolol, or clonidine may also be considered when either mineralocorticoid receptor antagonists or amiloride is contraindicated or adverse events occur.
## Interventional options for resistant hypertension
Interventional approaches represent a novel potential addition on lifestyle interventions and pharmacological therapy for the management of resistant hypertension. The neurogenic mechanisms implicated in BP elevation have been the target of several interventional approaches such as renal sympathetic denervation (RSD) and carotid baroreceptor activation therapy . The resistance of the arterial tree walls is another important factor contributing to the rise of BP. Recently, the creation of arteriovenous anastomosis between the distal iliac vein and artery to add a low resistance compartment to the arterial tree was investigated in patients with resistant hypertension 25 .
## Renal sympathetic denervation
The early RSD studies SYMPLICITY HTN-1 26 and -2 27 showed impressive reductions in BP and created high expectations for the future of the procedure in the hypertension treatment field. However, the first randomized sham-controlled study (SYMPLICITY HTN-3) failed to show any significant benefits of RSD over sham control in the reduction of both office and ambulatory BP in patients with drug-resistant hypertension 28 , and the technique fell into hibernation. Although several pitfalls were identified, the incomplete denervation and lack of circumferential, four-quadrant sympathetic fiber interruption were the main factors pointed out as causes of the negative findings of the study 29 . Three recent randomized sham-controlled studies used improved technologies and techniques to achieve optimal renal denervation and offered encouraging results.
The first one was the SPYRAL HTN-OFF MED trial, a randomized sham-controlled study of patients with untreated hypertension. Patients were randomly assigned to RSD of all accessible renal arterial vessels with the Symplicity Spyral multi-electrode catheter or the Symplicity G3 S (n = 38) or a sham procedure (n = 42). After 3 months, RSD resulted in significantly greater reductions in office and ambulatory BP levels compared with those of the sham control group (BP reduction differences of −7.7/−4.9 mm Hg and −5.0/−4.4 mm Hg in office and ambulatory BP, respectively). Importantly, no patient reported any safety concerns 30 . Similar results were observed in the SPYRAL HTN-ON MED trial, a randomized sham control study (of patients with uncontrolled hypertension on one to three anti-hypertensive drugs) that used the same techniques described for SPYRAL-OFF MED. RSD resulted in a greater reduction in office and ambulatory BP compared with the sham technique group (BP reduction differences of −6.8/−3.5 mm Hg and −7.4/−4.1 mm Hg in office and ambulatory BP, respectively). No reports of renal artery stenosis or worsening of renal function were reported 31 . The RADIANCE-HTN SOLO study was the third study that examined whether RSD performed with endovascular ultrasound reduces ambulatory BP in patients with hypertension in the absence of anti-hypertensive medication. After 2 months, significantly greater decreases in BP of −6.5/−4.1, −4.1/−1.8, and −7.1/−3.6 mm Hg in office, ambulatory, and home BP were reported with RSD compared with the sham control group 32 .
A recent study assessed the efficacy of different ablation methods in reducing BP 33 . Particularly, patients with resistant hypertension were assigned to either treatment with radiofrequency RSD of the main renal arteries, side branches, and accessories or an endovascular ultrasound-based RSD of the main renal artery. After 3 months, BP levels were significantly more reduced in the ultrasound ablation group compared with the radiofrequency ablation group of the main renal artery (−13.2 ± 13.7 versus −6.5 ± 10.3 mm Hg, respectively). Importantly, no significant difference was found between the radiofrequency ablation groups (−8.3 ± 11.7 mm Hg in the additional side branch ablation) or between the ultrasound and the side branch ablation groups, suggesting potential superiority of ultrasound over radiofrequency ablation of the main artery in reducing BP levels 33 . However, larger studies are needed to confirm or dispute the superiority of one technique over the other. Up to then, no approach could be considered the preferred or first-choice option.
Collectively, these studies resurfaced the clinical and research interest for RSD in the management of resistant hypertension. Although more patients demonstrated BP reduction with RSD than with the sham procedure, in several study participants who underwent RSD, an increase in BP was noticed, suggesting that some patients are not eligible for RSD. The quest of patient eligibility requires the identification of reliable and accurate predictors of BP response, a very demanding step with unknown outcome. Studies with larger study populations and longer follow-up periods are needed to establish the safety and efficacy of the technique, and several such studies are either planned or being conducted. The results of these studies will either re-enforce the concept of RSD or put the final nail in the coffin of this interventional approach.
## Iliac vein and artery anastomosis
The ROX CONTROL HTN study randomly assigned 83 patients with resistant hypertension to either pharmacological treatment plus placement of an arteriovenous coupler or pharmacological therapy alone. After 6 months, significant reductions in office and ambulatory systolic BP of 26.9 and 13.5 mm Hg were noted with the anastomosis device group, respectively. In contrast, such benefits were not observed in the control group (3.7 and 0.5 mm Hg, respectively). Similarly, a significant reduction in diastolic BP was reported with the device. However, implantation of the arteriovenous coupler was associated with ipsilateral venous stenosis in 29% of the patients, who received venoplasty or stenting 34 . After 12 months, office BP and ambulatory BP were reduced by 25/20.8 and 12.6/15.3 mm Hg, respectively, suggesting that the technique might offer long-lasting benefits in BP levels. Nevertheless, the percentage of patients who presented venous stenosis increased to 33%, who were treated successfully with venous stenting 35 . In conclusion, although iliac anastomosis showed promise in terms of efficacy, the safety concerns were significant and this approach was recently abandoned.
## Carotid baroreceptor activation therapy
Carotid baroreceptor activation therapy is a device-based approach aiming to activate the baroreceptors that signal the brain to activate a sympatholytic response. Such approaches might be useful in conditions characterized by sympathetic overactivity, such as hypertension, heart failure, and arrhythmias 36,37 . The potential benefits might be due to the reduction of heart rate (and thus cardiac workload and energy demands) and the manifestation of arterial dilation, which results in reduction of the peripheral resistance and enhancement of renal blood flow and natriuresis.
The Rheos Pivotal Trial assessed the impact of the Rheos system, a device that uses electrical impulses from a pulse generator to chronically activate the baroreflex at the carotid sinus, on BP levels in 265 patients with resistant hypertension. In the first group of patients, treatment was applied for the first 6 months, whereas in the second group, a delayed treatment initiation was implemented at the 6-month visit; 42% of patients in the first group versus 24% in the second group achieved systolic BP of less than 140 mm Hg at 6 months, and more than 50% in both groups had a systolic BP of less than 140 mm Hg at 12 months. However, the procedural safety endpoints of the study were not met since procedural complications occurred in 25% of the patients (transient or permanent nerve injury or general surgical complications) 38 .
However, the same company developed a second-generation device of smaller size, the Barostim Neo, which uses a smaller electrode on the surface of only one of the sinuses, thus reducing the invasiveness of the procedure and extending the battery life and replacement period 39 . The Barostim Neo System in the Treatment of Heart Failure/Barostim Hope for Heart Failure and the recent Baroreflex Activation Therapy in Patients with Heart Failure and a Reduced Ejection Fraction trials 40,41 demonstrated significant benefits in the functional status, quality of life, and exercise capacity in patients with heart failure and reduced injection fraction. Importantly, these studies met the safety endpoints; thus, the device was granted approval by the US Food and Drug Administration 39 . The device has also been approved in Europe for the management of resistant hypertension; the Barostim Neo trial demonstrated a persistent reduction in BP after 6 months (of a systolic BP of approximately 26 mm Hg) and an adequate safety profile 39,42 .
Overall, baroreflex activation therapy is approved for the treatment of resistant hypertension in Europe and for the treatment of heart failure with reduced ejection fraction in the US. However, it is more invasive than RSD, the safety of the procedure is not unequivocally proven, and thus it has not gained either wide application or general acceptance by hypertensive experts for the management of resistant hypertension.
The MobiusHD carotid bulb expansion device is an underexamination device that is used to reduce BP through stretching of the carotid artery at the bulb, which in turn activates the carotid baroreceptors. The Controlling and Lowering Blood Pressure with the MOBIUSHD (CALM-FIM_EUR) study recently reported pronounced reduction in BP levels of 24/12 mm Hg in office and 21/12 mm Hg in ambulatory BP with the use of the MobiusHD device in patients with resistant hypertension. Importantly, the device demonstrated an acceptable safety profile. Two other trials-the CALM-FIM_US 44 and the Controlling and Lowering Blood Pressure with the MobiusHD (CALM-2)studies-are examining the use of the MobiusHD device in patients with resistant hypertension. The results of these studies are eagerly awaited to further clarify the efficacy and safety of this approach and strengthen its role in the management of resistant hypertension.
## Continuous positive airway pressure therapy and resistant hypertension
Obstructive sleep apnea (OSA) is highly prevalent in patients with resistant hypertension . It has been suggested that the increased fluid retention and consequent upper airway edema may explain the high prevalence of OSA in these patients 52,53 . In addition, central fluid accumulation during sleep seems to significantly contribute to the manifestation and worsening of OSA 54-56 . Treatment of OSA with continuous positive airway pressure (CPAP) in patients with resistant hypertension was found to induce a modest but significant reduction in BP levels. In a study of patients with resistant hypertension and OSA, CPAP treatment resulted in a reduction in ambulatory BP of 3.1/3.2 mm Hg, an effect that was even greater in patients more adherent to CPAP therapy (reduction in ambulatory BP of 4.4/4.1 mm Hg with at least 4 hours of CPAP treatment per night). In contrast, a favorable impact of CPAP treatment on the prevention of cardiovascular events has not yet been demonstrated 58 .
## Refractory hypertension
The term refractory hypertension has been recently re-introduced and was included in the 2017 American guidelines for the management of hypertension 59 . Refractory hypertension is defined as failure of BP control with the use of five or more anti-hypertensive drugs of different drug classes, including a long-acting thiazide diuretic, such as chlorthalidone, and a mineralocorticoid receptor antagonist 59 . This novel type of uncontrolled hypertension seems to be rare, affecting less than 5% of the patients referred to a specialized clinic for uncontrolled hypertension. Furthermore, refractory hypertension was found to be more frequent in African-American, younger, and female patients . There is evidence suggesting that the cause of treatment failure in patients with refractory hypertension, in contrast to resistant hypertension, is the increased sympathetic tone, rather than fluid retention, as indicated by the increased heart rate levels and urine norepinephrine excretion in such patients . Therefore, intensification of diuretic therapy may not be effective, and sympatholytic agents or device-based therapy may be preferable . However, data in such patients are still missing and studies are needed to identify optimal management options.
# Conclusions
During the past few years, important data for the management of resistant hypertension have emerged. Most data support that mineralocorticoid antagonists (and especially spironolactone) present more favorable BP-lowering properties in patients with resistant hypertension compared with central acting drugs, alpha-and beta-blockers. In case of contraindications or adverse events, amiloride should be used as an alternative option followed by doxazosin, bisoprolol, or clonidine.
Several device-based approaches are being investigated, and recent RSD trials have rekindled interest in the interventional therapy of resistant hypertension. The few studies implementing carotid baroreceptor stimulation have shown favorable results, which need to be verified in controlled trials with a long follow-up period, while safety concerns need to be adequately addressed. Iliac anastomosis devices are no longer available in our therapeutic armamentarium. Overall, we are living in exciting times in the resistant hypertension field, and a lot of data, especially about the role of interventional approaches in the treatment of resistant hypertension, are eagerly expected in the near future. |
Initiation of wound healing is regulated by the convergence of mechanical and epigenetic cues
Wound healing in the skin is a complex physiological process that is a product of a cell state transition from homeostasis to repair. Mechanical cues are increasingly being recognized as important regulators of cellular reprogramming, but the mechanism by which it is translated to changes in gene expression and ultimately cellular behavior remains largely a mystery. To probe the molecular underpinnings of this phenomenon further, we used the downregulation of caspase-8 as a biomarker of a cell entering the wound healing program. We found that the wound-induced release of tension within the epidermis leads to the alteration of gene expression via the nuclear translocation of the DNA methyltransferase 3A (DNMT3a). This enzyme then methylates promoters of genes that are known to be downregulated in response to wound stimuli as well as potentially novel players in the repair program. Overall, these findings illuminate the convergence of mechanical and epigenetic signaling modules that are important regulators of the transcriptome landscape required to initiate the tissue repair process in the differentiated layers of the epidermis.ResultsWound induced down-regulation of caspase-8 RNA correlates with the degree of promoter methylationPreviously, we have established the importance of the down-regulation of caspase-8 RNA in both physiological (wound healing[12]) as well as pathological (atopic dermatitis[15]andPLOS BIOLOGYRegulation of the wound-healing response PLOS Biology | https://doi.
# Introduction
The wound healing program in an epithelial tissue is fundamentally a product of cell state transitions from homeostasis to a repair program. In particular, cutaneous wound healing in the adult is an intricately regulated system wherein keratinocytes and many other cell lineages exhibit their plasticity as they undergo reprogramming, to carry out otherwise dormant functions, to rebuild the damaged skin. Many of the phenomena that occur in the repair process in adult skin are, in fact, reminiscent of cellular events that operate during fetal development [bib_ref] Wound repair: A showcase for cell plasticity and migration, Shaw [/bib_ref]. At the other extreme, inappropriate activation of these repair processes can manifest as tissue pathology, which forms the foundation of the perception of diseases with a "wound signature" [bib_ref] Activation of Host Wound Responses in Breast Cancer Microenvironment, Troester [/bib_ref]. The question that arises is how the whole scale changes in gene expression are accomplished in order to facilitate this cellular reprogramming.
Recently, epigenetic regulators have emerged as a vital component capable of transiently rewiring the cell's transcriptional program to mediate the continual regeneration of the mouse epidermis [bib_ref] The Epigenetic Regulation of Wound Healing, Lewis [/bib_ref] [bib_ref] Dnmt3a and Dnmt3b Associate with Enhancers to Regulate Human Epidermal Stem Cell..., Rinaldi [/bib_ref]. This mode of gene regulation operates at multiple levels ranging from histone and DNA modifications, chromatin remodeling, and activity of various subtypes of RNA species such as non-coding RNAs and micro-RNAs (miRNAs) [bib_ref] Epidermal Stem Cells and Their Epigenetic Regulation, Shen [/bib_ref] [bib_ref] Epigenetic Regulation of Skin Cells in Natural Aging and Premature Aging Diseases, Orioli [/bib_ref]. These epigenetic mechanisms can thus have a profound impact on the transcriptional landscape of the cell and can easily be envisioned to participate in the transient activation or repression of approximately 1,000 genes that are required for wound closure [bib_ref] Wound healing and inflammation genes revealed by array analysis of "macrophageless" PU.1..., Cooper [/bib_ref]. Circumstantial evidence in support of a role for epigenetics in tissue repair comes from reports of the dynamic expression of several epigenetic regulators following injury to the skin. For instance, Ezh2, Suz12, and Eed, which are components of the polycomb repressive complex 2 (PRC2), are down-regulated, whereas the histone methylases JMJD3 and Utx are up-regulated upon tissue damage and all return to homeostatic levels upon the completion of wound closure [bib_ref] Epigenetic reprogramming during wound healing: loss of polycomb-mediated silencing may enable upregulation..., Shaw [/bib_ref]. While the description of various epigenetic players in epidermal homeostasis and wound healing are reported, the identity and function of their upstream regulators are, to a large extent, absent in the literature.
An intriguing candidate for an upstream regulator in a highly tensile tissue such as the epidermis are mechanical cues. The epidermis is a stratified epithelium comprised of a basal layer of proliferation competent keratinocytes and suprabasal layers of differentiated cells glued together via intercellular adhesion complexes that partly endows the tissue with its barrier function. In different cell types, changes in mechanical tension have been documented to induce the nuclear translocation of important transcription factors-a notable example of which is YAP/TAZ that has proliferation stimulating gene targets [bib_ref] YAP/TAZ upstream signals and downstream responses, Totaro [/bib_ref]. Many studies, including those on epidermal homeostasis and wound healing, have primarily focused on the changes in gene expression in proliferating cells [bib_ref] Epidermal stem cells in wound healing and their clinical applications, Yang [/bib_ref] [bib_ref] Epidermal Stem Cells in Homeostasis and Wound Repair of the Skin, Senoo [/bib_ref]. On the other hand, differentiated cells, such as the suprabasal keratinocytes near the surface of the epidermis, have largely been relegated to bystander status. In spite of this, a few reports suggest that these neglected pools of differentiated cells are not inert in the cellular crosstalk that mediates the early responses of the tissue to injury. In particular, the uppermost layer of differentiated keratinocytes in the epidermis expresses caspase-8 that has a non-canonical role in regulating the wound healing program. We previously demonstrated that the down-regulation of caspase-8 is a natural phenomenon upon application of an excisional wound to the mouse skin [bib_ref] Dynamic expression of epidermal caspase 8 simulates a wound healing response, Lee [/bib_ref]. This down-regulation is particularly relevant as genetically ablating caspase-8 in the epidermis is sufficient to induce a wound healing response even in the absence of any damage to the organ. In addition, the down-regulation of caspase-8 in the upper, differentiated layer of the epidermis mediates signaling networks to incite epithelial stem cell proliferation in the epidermis [bib_ref] Dynamic expression of epidermal caspase 8 simulates a wound healing response, Lee [/bib_ref] and the hair follicle [bib_ref] Stimulation of hair follicle stem cell proliferation through an IL-1 dependent activation..., Lee [/bib_ref] to fuel wound closure. We have thus used the down-regulation of caspase-8 as a cellular biomarker to identify the higher order regulatory machinery that reprograms the cell to enter the wound healing process in differentiated keratinocytes, which are emerging as an important participant in the tissue repair program.
psoriasis [bib_ref] Sustained Secretion of the Antimicrobial Peptide S100A7 Is Dependent on the Downregulation..., Bhatt [/bib_ref] scenarios. The mechanisms responsible for this down-regulation, however, remain unknown. Uncovering the regulatory machinery of caspase-8 RNA also holds the promise of understanding the process by which cells transition from a state of homeostasis to repair. Moreover, it can provide potential new therapeutic targets for common inflammatory skin diseases where this regulation is perturbed.
RNA down-regulation can be achieved either via blocking the synthesis and/or active degradation. In order to distinguish between these 2 possibilities, we determined the half-life of caspase-8 in homeostasis compared to wound conditions. In differentiated primary epidermal keratinocytes, we observed that the half-life of caspase-8 mRNA under homeostatic conditions in vitro is approximately 2 hours (S1A In an in vitro scratch wound assay with multiple scratches, the level of caspase-8 RNA is significantly reduced by 8 hours [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref]. Since the reduction of caspase-8 is faster under homeostatic conditions compared to the wound healing context, merely blocking RNA synthesis can achieve the reduction of caspase-8 mRNA and initiate the downstream wound healing response. Interestingly, the reduction caspase-8 RNA is localized in cells near the front of the scratch wound in vitro (Figs 1B and S1B). In situ hybridization of caspase-8 RNA demonstrates that the down-regulation can clearly be visualized in the cells immediately adjacent to the leading edge of a single scratch wound as early as 4 hours post wounding. By 8 hours post wounding, the caspase-8 RNA is down-regulated in about 3 to 4 cell rows from the wound front. These findings are consistent with our observation in excisional wounds on the back skin of mice where the decrease of caspase-8 RNA is visible as early as 4 hours in the wound proximal region (Figs 1C and S1C). Together, these results suggest that simply blocking transcription post injury is sufficient to down-regulate caspase-8. We hypothesized that the block in caspase-8 RNA synthesis is achieved through promoter methylation, which is consistent with previous reports documenting the same phenomenon in a variety of cancer cells through the hypermethylation of regulatory DNA sequence [bib_ref] Caspase-8 in cancer biology and therapy, Fulda [/bib_ref] [bib_ref] Caspase-8 as a therapeutic target in cancer, Stupack [/bib_ref]. To understand whether this process in cancer cells is an aberration of the physiological healing program, we have assessed the methylation status of important regulatory sequences in the caspase-8 promoter, namely the CpG loci and SP1 binding sites (S1D Fig) [bib_ref] Silencing of caspase-8 in murine hepatocellular carcinomas is mediated via methylation of..., Liedtke [/bib_ref]. Analysis of methylation of SP1 sites and other CpG loci reveals a time-dependent increase of promoter methylation in a sheet of differentiated epidermal keratinocytes subjected to multiple scratch wounds [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref]. This progressive increase in the methylation of the caspase-8 promoter correlates well with the kinetics of the decrease in caspase-8 RNA [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref]. This suggests DNA methylation may play a critical role in regulating the wound healing response.
## Wound stimuli induce the nuclear localization of the dna methyltransferase dnmt3a
We thus investigated the mechanism responsible for DNA methylation of the caspase-8 promoter in response to injury. The bisulfite sequencing data reveals that the methylation of the caspase-8 promoter is a de novo event in response to wounding. We therefore examined the status of the 2 known de novo DNA methyltransferases (DNMTs), namely DNMT3a and DNMT3b, in response to injury. Interestingly, de novo DNMTs (DNMT3a and 3b) have also been shown to be important in regulating epidermal stem cell homeostasis [bib_ref] Dnmt3a and Dnmt3b Associate with Enhancers to Regulate Human Epidermal Stem Cell..., Rinaldi [/bib_ref]. To investigate whether these enzymes likewise play a role in tissue repair, we examined their expression in the wounded epidermis. Consistent with a previous report, under homeostatic conditions, we found that DNMT3a mainly resides in the nucleus of the basal/proliferating (K5 positive) cells and is absent or cytoplasmic in the suprabasal/differentiated (K5 negative) keratinocytes (S2A and S2B [bib_ref] Loss of Dnmt3a and Dnmt3b does not affect epidermal homeostasis but promotes..., Rinaldi [/bib_ref]. This localization was also recapitulated in vitro wherein we observed the Interestingly, in vivo we observed that DNMT3a undergoes cytoplasmic to nuclear translocation in cells adjacent to the wound [fig_ref] Fig 2: De novo DNMT3a increases nuclear localization at wound proximal site [/fig_ref]. Quantification of the nuclear versus cytoplasmic localization of DNMT3a revealed a time-dependent accumulation of the enzyme in the nucleus post wounding [fig_ref] Fig 2: De novo DNMT3a increases nuclear localization at wound proximal site [/fig_ref]. This phenomenon was more apparent in an in vitro scratch assay, where keratinocytes adjacent to the scratch exhibited nuclear localization of DNMT3a [fig_ref] Fig 2: De novo DNMT3a increases nuclear localization at wound proximal site [/fig_ref]. The second known de novo DNMT, DNMT3b, also showed cytoplasmic localization in differentiated keratinocytes (S2D However, it did not translocate to the nuclei of wound proximal keratinocytes (S2E .
Thus, we focused on understanding the mechanistic details of DNMT3a's role in regulating wound healing program. The increase in DNMT3a nuclear localization was time dependent, affecting wound proximal keratinocytes first and then propagates toward distal cells. At the completion of the wound healing program, we observe that DNMT3a localization is again prominent within the cytoplasms of differentiated (K5 negative) keratinocytes, while nuclear localization is restricted to cells in the basal layer of the epidermis (S2F In conclusion, we observe that the DNMT3a shows significant nuclear localization in the wound-proximal (leading edge) cells within 4 hours of the injury and the localization pattern further penetrates in the distal regions as time passes [fig_ref] Fig 2: De novo DNMT3a increases nuclear localization at wound proximal site [/fig_ref]. The nuclear localization kinetics also correlates with the pattern of caspase-8 down-regulation as well as promoter methylation [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref].
## Dnmt3a directly regulates caspase-8 expression
We further explored whether the de novo DNA methylation of caspase-8 promoter is the result of DNMT3a's direct binding to this region (S1B . This was accomplished with the use of chromatin immunoprecipitation (ChIP) to assess the level of DNMT3a occupancy on the caspase-8 promoter pre-and post-scratch wound. We found that scratch wounds lead to the higher occupancy of DNMT3a on caspase-8 promoter, which is not seen in the case of DNMT3b . To understand the functional relevance of DNMT activity in maintaining caspase-8 levels, we pre-treated the differentiated keratinocytes with a generic DNMT inhibitor (5-Aza-2 0 -deoxycytidine). We observed that the inhibitor treated cells were unable to down-regulate caspase-8 mRNA in a scratch wound assay (S3A To specifically assess the role of DNMT3a, we performed shRNA-mediated knockdown of DNMT3a (S3B Compared to the controls, keratinocytes with reduced DNMT3a expression were unable to downregulate caspase-8 in response to scratch wound . We further analyzed whether failure of caspase-8 mRNA down-regulation was due to the absence of promoter methylation. Indeed, scratch wounded keratinocytes, transduced with DNMT3a shRNA, showed significantly reduced DNA methylation pattern on the caspase-8 promoter compared to scrambled RNA control .
Promoter activities are often dependent on the associated histone modifications. These histone marks generally guide the DNA methylation at a particular genic region and vice-a-versa [bib_ref] Lateral Thinking: How Histone Modifications Regulate Gene Expression, Lawrence [/bib_ref] [bib_ref] DNA methylation pathways and their crosstalk with histone methylation, Du [/bib_ref] [bib_ref] Reversible Regulation of Promoter and Enhancer Histone Landscape by DNA Methylation in..., King [/bib_ref]. DNMT3a occupancy and activity has also been shown to be influenced by the methylation status of certain lysine (K) residues on the histone 3 (H3) tail [bib_ref] DNA methylation pathways and their crosstalk with histone methylation, Du [/bib_ref] [bib_ref] Structural insight into autoinhibition and histone H3-induced activation of DNMT3A, Guo [/bib_ref]. To investigate the core machinery required for DNMT3a-mediated methylation on the caspase-8 promoter, we assessed several activation and repression histone marks in scratch wounded keratinocytes . We observed that 2 transcriptional activation marks, H3K9ac and H3K4me3, are decreased at the caspase-8 promoter. On the other hand, the H3K9me3 mark, which is are shown as mean ± SEM, P-values were calculated using 1-way ANOVA with Dunnett's test and 2-tailed t test (A), ��� P � 0.001, ns = P > 0.05). Data underlying the graphs can be found in [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref] associated with transcriptional repression, was significantly increased at the caspase-8 promoter following wounding. Interestingly, another classical repressive mark, H3K27me3, did not show a significant change. It is possible that the caspase-8 proximal promoter is another example of a bivalent promoter [bib_ref] A double take on bivalent promoters, Voigt [/bib_ref] having both activation (H3K9ac and H3K4me3) and repression (H3K27me3) marks. In this scenario, then, wound-mediated repression of caspase-8 is achieved via reduction of both H3K9ac and H3K4me3 along with an increase in the H3K9me3 mark and DNMT3a occupancy. These results establish the mechanism by which DNMT3a localizes to the caspase-8 promoter. An outstanding question is whether DNMT3a is required for a proper wound healing response. To address this issue, we tested the effect of the knockdown of DNMT3a in a scratch wound assay . We found that keratinocytes with decreased DNMT3a exhibited an impaired wound closure response, thereby illustrating the necessity of this methyltransferase in the proper repithelialization of an in vitro wound.
## Cellular tension mediates dnmt3a localization and caspase-8 expression
We observed that caspase-8 down-regulation and DNMT3a nuclear localization initiate at the edge of wound site [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref]. Given that these are early responses to injury, understanding the mechanistic basis of this phenomenon can provide insights into the broader process of cellular wound sensing. The keratinocytes in the epithelial sheet are strongly adhered to each other and an event of injury will result in the sudden relaxation in that tension, particularly in the cells at the boundary of the wound. Interestingly, the expanding number of cells exhibiting the down-regulation of caspase-8 RNA in the scratch wound assay over time [fig_ref] Fig 1: Kinetics of caspase-8 promoter methylation and expression [/fig_ref] closely parallels the changes in traction force previously reported for the collective cell migration of an epithelial sheet following a scratch wound [bib_ref] Physical forces during collective cell migration, Trepat [/bib_ref]. We therefore investigated whether release of tension, caused by the severing of the epithelial sheet, can impact DNMT3a subcellular localization and subsequently caspase-8 expression. As shown in S4A in cellular tension can be achieved via targeting the components of the adherens junction, which are known to play a role in generating and maintaining cellular tension [bib_ref] E-cadherin junctions as active mechanical integrators in tissue dynamics, Lecuit [/bib_ref] [bib_ref] Cadherin Adhesion and Mechanotransduction, Leckband [/bib_ref].
We observed that tension release by disrupting calcium-dependent E-cadherin junctions via EGTA treatment resulted in the nuclear localization of DNMT3a (Figs 4A and S4B). Similarly, releasing cellular tension endowed by nonmuscle myosin II (NM-II) with the pharmacological inhibitor of NMII, blebbistatin, induced the DNMT3a's nuclear translocation from the cytosol [fig_ref] Fig 4: Effect of cellular tension on DNMT3a localization and caspase-8 expression [/fig_ref]. Furthermore, we examined the effect of blocking release of cellular tension in a scratch wounded sheet of epidermal keratinocytes. The release of tension was blocked by pre-treating keratinocytes with calyculin-A, which inhibits myosin light-chain phosphatase, thereby maintaining the active state of NMII [bib_ref] RhoA is dispensable for skin development, but crucial for contraction and directed..., Jackson [/bib_ref]. The treatment of keratinocytes with calyculin-A prior to scratch wounding blocked the nuclear translocation of DNMT3a that was observed in cells treated with vehicle control [fig_ref] Fig 4: Effect of cellular tension on DNMT3a localization and caspase-8 expression [/fig_ref].
In addition to a pharmacological approach, we also modulated cellular tension by altering the substrate stiffness on which the keratinocytes were growing. This was accomplished by utilizing polyacrylamide gels of various stiffness, which would alter cellular tension. We observed that differentiated keratinocytes seeded on "soft" matrices ranging from 10 kPa to 40 kPa mostly harbored DNMT3a in the nuclei [fig_ref] Fig 4: Effect of cellular tension on DNMT3a localization and caspase-8 expression [/fig_ref]. However, cells grown on a "stiffer" matrix (100 kPa) predominantly showed a cytoplasmic localization of DNMT3a. We then evaluated whether DNMT3a's dynamic localization in response to pharmacological and mechanical alterations in cellular tension has any transcriptional consequences. We observed that in all the scenarios where DNMT3a nuclear localization was favored (scratch wounds, EGTA/blebbistatin treatment, soft substrates), caspase-8 RNA was down-regulated compared to their respective controls [fig_ref] Fig 4: Effect of cellular tension on DNMT3a localization and caspase-8 expression [/fig_ref]. On the other hand, inhibition of DNMT3a's nuclear localization (via calyculin-A or a stiff substrate) resulted in the failure of caspase-8 down-regulation in spite of a scratch wound. These results suggest a correlation between DNMT3a localization and changes in tensile forces. It should be noted that these interventions may have other effects on the cell in addition to modulating cellular tension, and thus we cannot rule out additional pathways leading to cellular reprograming via epigenetic means.
## Dna methylation could be a global regulator of gene expression to initiate wound healing program
We further assessed whether the down-regulation of caspase-8 is a paradigm for the global down-regulation of genes to achieve a cell state transition from homeostasis to wound healing. Surprisingly, the transcriptome profile of scratch wounded differentiated keratinocytes has not been reported even though these layers are the first to encounter damage in vivo. Thus, we performed RNA sequencing of wounded v/s unwounded primary mouse keratinocyte that were differentiated via the calcium switch protocol (S5A The analysis of the transcriptome data revealed that the number of down-regulated genes outnumbered the up-regulated genes post injury. We verified the sequencing data by specifically analyzing genes via qPCR that have already been implicated in wound healing or epidermal development Interestingly, there was an inverse correlation with the RNA expression and the degree of methylation for many of the genes we interrogated. This suggest that DNA methylation could be a global regulator for a set of wound response genes (in addition to caspase-8) needed for the wound healing program.
Analysis of transcriptome data has revealed many such group of genes and their biological processes (S6 Of particular interest were the down-regulation of genes involved in the differentiation of keratinocytes. The Ghazizadeh lab has reported evidence of dedifferentiation of suprabasal keratinocytes as a mode of aiding cutaneous regeneration and repair. Interestingly, the regeneration of skin epithelia by differentiated epidermal cultures was found to be facilitated by the capacity of these cells to proliferate [bib_ref] Regeneration of multilineage skin epithelia by differentiated keratinocytes, Mannik [/bib_ref]. The transcriptome profile of scratch wounded differentiated keratinocytes reveals an up-regulation of cell cycle associated genes that is consistent with this report. Consequently, the convergence of mechanical and epigenetic cues appears to play an important role in the plasticity of differentiated epidermal keratinocytes in cutaneous repair and regeneration. The processes that occur during the wound healing phases of inflammation, proliferation, and tissue remodeling are often recapitulated in a deregulated manner in many pathologies leading to the notion of diseases with a "wound signature." Prominent among these is the view of cancer as an over healing wound [bib_ref] Cancer as an overhealing wound: an old hypothesis revisited, Schä Fer [/bib_ref]. As we noted earlier, there is a body of literature demonstrating that the down-regulation of caspase-8 in cancer cells is accompanied with the methylation of its promoter region [bib_ref] Promoter methylation pattern of caspase-8, P16INK4A, MGMT, TIMP-3, and E-cadherin in medulloblastoma, Ebinger [/bib_ref] [bib_ref] Caspase 8 and maspin are downregulated in breast cancer cells due to..., Wu [/bib_ref] [bib_ref] Epigenetic methylation and expression of caspase 8 and survivin in hepatocellular carcinoma, Cho [/bib_ref] [bib_ref] Impact of the DNA methyltransferases expression on the methylation status of apoptosis-associated..., Hervouet [/bib_ref]. In addition, we have previously demonstrated that inflammatory human skin diseases such as atopic dermatitis scratch wounded keratinocytes, transduced with either scrambled RNA or DNMT3a shRNA. (D) ChIP-qPCR analysis of H3K9ac, H3K4me3, H3K9me3, and H3K27me3 at caspase-8 promoter in control and scratch wounded keratinocytes (n = 3). (E) Effect of DNMT3a down-regulation on in vitro wound healing assay (n = 3). (Data are shown as mean ± SEM, P-values were calculated using 2-tailed t test (A, B, D), � P � 0.05, �� P � 0.01, ��� P � 0.001, ns = P > 0.05.) Data underlying the graphs can be found in , and 3E of S1 Raw Data. ChIP, chromatin immunoprecipitation; DNMT3a, DNA methyltransferase 3A.
https://doi.org/10.1371/journal.pbio.3001777.g003 [bib_ref] Development of atopic dermatitis-like skin disease from the chronic loss of epidermal..., Li [/bib_ref] and psoriasis [bib_ref] Sustained Secretion of the Antimicrobial Peptide S100A7 Is Dependent on the Downregulation..., Bhatt [/bib_ref] likewise exhibit a loss of epidermal caspase-8. To probe a possible link between caspase-8 down-regulation and methyltransferase expression, we utilized the imiquimod-induced model of psoriasis in mice. In the psoriatic skin of mice, we observed robust nuclear localization of DNMT3a in all the epidermal layers, whereas in the control animals, nuclear DNMT3a was primarily localized in the basal keratinocytes Altogether, this suggests that the epigenetic regulation governing the cell state transition in wound healing is usurped in many diseases ranging from inflammatory skin diseases to carcinomas.
# Discussion
The wound healing literature involving epidermal keratinocytes have elegantly described many signaling pathways and gene expression profiles in the proliferating cells of the basal layer [bib_ref] Epithelialization in Wound Healing: A Comprehensive Review, Pastar [/bib_ref] [bib_ref] Numerous keratinocyte subtypes involved in wound re-epithelialization, Patel [/bib_ref]. In contrast, differentiated epidermal cells, such as the suprabasal keratinocytes near the outer surface of the skin, have largely been overlooked for their potential role during wound healing. Interestingly, our previous work demonstrates that the uppermost layer of differentiated keratinocytes, namely the granular layer, expresses caspase-8 that has a non-canonical role in regulating the wound healing program [bib_ref] Dynamic expression of epidermal caspase 8 simulates a wound healing response, Lee [/bib_ref]. It was found that the down-regulation of caspase-8 is both necessary and sufficient to induce a wound healing response in the absence of any tissue damage. In addition, the chronic down-regulation of caspase-8 underlies inflammatory skin diseases such as atopic dermatitis [bib_ref] Development of atopic dermatitis-like skin disease from the chronic loss of epidermal..., Li [/bib_ref] and psoriasis [bib_ref] Sustained Secretion of the Antimicrobial Peptide S100A7 Is Dependent on the Downregulation..., Bhatt [/bib_ref]. These findings have made the decrease in caspase-8 expression a useful wound healing biomarker and led us to inquire about the mechanism of caspase-8 regulation in skin keratinocytes.
Clues about the regulation of caspase-8 are reported in the context of cancer. Similar to the wound healing process, it is generally down-regulated in various cancers [bib_ref] Caspase-8 as a therapeutic target in cancer, Stupack [/bib_ref] [bib_ref] High levels of expression and nuclear localization of interleukin-1 beta converting enzyme..., Nakagawara [/bib_ref] [bib_ref] A Novel Target for Prevention and Therapy, Subramaniam [/bib_ref]. It is possible that the cancers, known as "over healing wound," usurp physiological pathway of wound healing for its own propagation [bib_ref] Cancer as an overhealing wound: an old hypothesis revisited, Schä Fer [/bib_ref]. Here, we show that wound sensing leads to the acute increase in the caspase-8 promoter methylation as a potential mechanism of gene silencing. This parallels with the findings on caspase-8 down-regulation in hepatocellular carcinoma, where methylation status of SP1 sites and nearby CpG dinucleotides in the promoter region were proposed to be a major regulator of caspase-8 expression [bib_ref] Silencing of caspase-8 in murine hepatocellular carcinomas is mediated via methylation of..., Liedtke [/bib_ref]. In fact, it has been observed that caspase-8 and several other genes are known to be down-regulated in various cancers via DNA methyltransferase (DNMT) activity [bib_ref] Caspase-8 as a therapeutic target in cancer, Stupack [/bib_ref] [bib_ref] High levels of expression and nuclear localization of interleukin-1 beta converting enzyme..., Nakagawara [/bib_ref] [bib_ref] A Novel Target for Prevention and Therapy, Subramaniam [/bib_ref]. The overexpression of DNMT3a has also been shown to be associated with several cancers [bib_ref] DNMT3A/3B overexpression might be correlated with poor patient survival, hypermethylation and low..., He [/bib_ref] [bib_ref] DNMT3A overexpression is associated with aggressive behavior and enteroblastic differentiation of gastric..., Kataoka [/bib_ref]. The process of de novo DNA methylation during an acute physiological response such as wound healing is a rarely described phenomenon. Mammalian cells are known to have only 2 de novo DNA methyltransferases, DNMT3a and DNMT3b. Both have been widely studied for their role in physiological processes like embryogenesis [bib_ref] Synergistic Function of DNA Methyltransferases Dnmt3a and Dnmt3b in the Methylation of..., Li [/bib_ref] and hematopoiesis [bib_ref] Dnmt3a and Dnmt3b have overlapping and distinct functions in hematopoietic stem cells, Challen [/bib_ref] , as well as pathological conditions such as cancer [bib_ref] DNA methyltransferases and their roles in tumorigenesis, Zhang [/bib_ref] [bib_ref] DNA methylation, methyltransferases, and cancer, Robertson [/bib_ref]. In particular, it has been shown that DNMT3a and 3b are required as regulators of enhancer activity and RNA production of genes necessary for epidermal stem cell homeostasis [bib_ref] Dnmt3a and Dnmt3b Associate with Enhancers to Regulate Human Epidermal Stem Cell..., Rinaldi [/bib_ref]. In disease context, DNMT3a has been described to be overexpressed or mutated in various carcinomas [bib_ref] DNMT3a expression pattern and its prognostic value in lung adenocarcinoma, Husni [/bib_ref] [bib_ref] Epigenetic modifiers DNMT3A and BCOR are recurrently mutated in CYLD cutaneous syndrome, Davies [/bib_ref] and correlates with the down-regulation of caspase-8 in these same scenarios. Here, we found that upon injury to the skin or differentiated epidermal sheets, the suprabasal cells near wound edge showed a nuclear localization of DNMT3a, but not DNMT3b. We have captured that DNMT3a indeed occupies the caspase-8 promoter and plays an important role in its down-regulation post injury. Importantly, there are numerous additional promoters that also methylated and gene expression is using 2-tailed t test (D), � P � 0.05, �� P � 0.01, ��� P � 0.001, ns = P > 0.05). Data underlying the graphs can be found in [fig_ref] Fig 4: Effect of cellular tension on DNMT3a localization and caspase-8 expression [/fig_ref] transcriptionally down-regulated. This indicates that DNA methyltransferases have a broad spectrum of genomic targets that work in combination to fuel the cell state transition from homeostasis to repair. In parallel to the DNA methylation, the literature also describes changes in histone modifications responsible for the ON/OFF state of a particular gene. The histone modifications and their modifiers have been studied in depth to understand how the expressions of various epidermal differentiation genes are regulated [bib_ref] Epigenetic regulation of skin: Focus on the Polycomb complex, Zhang [/bib_ref]. In general, H3K9ac and H3K4me3 are considered as gene activation marks and H3K9me3 and H3K27me3 are known as repression marks. It is also observed that certain methylation state of H3K36 dictates the DNMT3a's recruitment to a particular DNA segment on the chromosome [bib_ref] Engineering of a Histone-Recognition Domain in Dnmt3a Alters the Epigenetic Landscape and..., Noh [/bib_ref] [bib_ref] The histone mark H3K36me2 recruits DNMT3A and shapes the intergenic DNA methylation..., Weinberg [/bib_ref]. In our efforts to understand the histone modifications during wound healing, we observed a reduction in H3K9ac and H3K4me3 levels, along with an increase in the H3K9me3 mark at the caspase-8 promoter. These histone modifications are known to be regulated via various other epigenetic players such as polycomb repressive complexes (PRC 1/2), JMJD, Setd8, and HDACs during epidermal development [bib_ref] Epigenetic regulation of skin: Focus on the Polycomb complex, Zhang [/bib_ref].
How these epigenetic players are regulated is another important question in the field. While there are many chemical cues, adhesion signals, and transcription factors described to regulate the wound healing process, emerging evidence links mechanical forces to epigenetic and transcriptional responses [bib_ref] Mechanotransduction in Wound Healing and Fibrosis, Kuehlmann [/bib_ref] [bib_ref] Epigenetic regulation and mechanobiology, Li [/bib_ref]. Even during the development of epidermal tissue, tension generating molecular players like nonmuscle myosin IIA (NMIIA), along with emerin (Emd) and PRC2 regulate the differentiation process of epidermal stem cells. The strain on epidermal cells reduces Emd levels from the inner nuclear membrane, which then leads to the loss of the histone mark H3K9me2,3. This is followed by PRC2 mediated increase of H3K27me3 occupancy at several heterochromatic regions and thereby gene silencing [bib_ref] Mechanical regulation of transcription controls Polycomb-mediated gene silencing during lineage commitment, Le [/bib_ref]. Along the same line, recently Nava and colleagues has described how short-and long-term mechanical stress on a cell can result in changes in stiffness of the nuclear membrane, loss of H3K9me3 marks at the heterochromatin, and overall chromatin and cytoskeletal reorganization [bib_ref] Heterochromatin-Driven Nuclear Softening Protects the Genome against Mechanical Stress-Induced Damage, Nava [/bib_ref]. These are some of the key discoveries suggesting external mechanical forces drive changes in heterochromatin organization, gene expression changes, and cytoskeletal reorganization in a way that mechanical energy gets redistributed and DNA damage can be avoided. In this context, our results demonstrate that the release in the mechanical tension, either by physical or chemical treatments, results in the DNMT3a's nuclear localization and down-regulation of caspase-8. This observation is consistent with the concept of mechano-sensitive histone modifications, which could lay a foundation for the occupancy of DNMT3a. In a wider context of cellular reprogramming during the wound response, mechanotransduction seem to have a large impact on the transcriptome of the cell via the concomitant initiation of several epigenetic pathways. Future studies in this area will include elucidation of the connection between the release of mechanical tension and their sensing by these epigenetic machineries. For example, DNMT3a has been shown to have multiple binding partners (DNMT3L, SUMO-1, Cbx4, Ubc9, RP58, HDAC1) for their nuclear shuttling as well as chromosomal occupancy, some of which can potentially function as a primary signal sensor to guide the localization of DNMT3a [bib_ref] Modification of de novo DNA methyltransferase 3a (Dnmt3a) by SUMO-1 modulates its..., Ling [/bib_ref] [bib_ref] Polycomb protein Cbx4 promotes SUMO modification of de novo DNA methyltransferase Dnmt3a, Li [/bib_ref]. Moreover, in different cell types, changes in mechanical tension have been documented to directly induce the nuclear translocation of important transcription factors. A notable example of which is the YAP/TAZ complex, which has proliferation stimulating gene targets [bib_ref] The Roles of YAP/TAZ and the Hippo Pathway in Healthy and Diseased..., Rognoni [/bib_ref].
The described model of mechanosensitive epigenetic players would obviously be regulating a larger gene regulatory network, in addition to caspase-8. Interestingly, the transcriptome literature on wound healing has utilized proliferating keratinocytes, leaving the transcriptome profile of differentiated keratinocytes unknown despite the fact that it constitutes about 2/3 of the epidermis. Our research fills an important gap by providing a transcriptome profile of in vitro wounded differentiated keratinocytes. The results give us a unique insight in the regulation of various unexplored wound-response genes. On a particular note, we observe a strong down-regulation of multiple epidermal differentiation genes in response to injury. From the current transcriptome and literature survey, it is evident that various keratinocyte differentiation markers (such as involucrin, keratins K1/K10, and filaggrin) are down-regulated along with cell adhesion molecules (involved in tight junction, adherens junctions, and desmosomes). This is consistent with a report from S. Ghazizadeh's lab that de-differentiation of suprabasal keratinocytes is a contributing factor in the wound healing response [bib_ref] Regeneration of multilineage skin epithelia by differentiated keratinocytes, Mannik [/bib_ref]. Our data suggest that the release of mechanical tension in differentiated keratinocytes is one component in this process by inducing a "partial de-differentiation" and perhaps additional soluble signaling cues are required to achieve complete dedifferentiation.
# Materials and method
# Ethics statement
## Cell culture and scratch wound assay
The isolation of primary keratinocytes from neonatal mice was performed as described in [bib_ref] Isolation and culture of primary mouse keratinocytes from neonatal and adult mouse..., Li [/bib_ref]. Briefly, mice pups were sacrificed and the skin was removed. The skin was kept in dispase at 4˚C overnight (or 37˚C for 1 hour) to separate epidermis. The epidermis was then digested with trypsin to isolate keratinocytes. These cells were filter with 70-μm mesh and cultured further as described in [bib_ref] Isolation and culture of epithelial stem cells, Nowak [/bib_ref]. The keratinocytes were cultured in lab with feeder cells (3T3J2) for 10 passages. Then, feeder-independent keratinocytes were taken and tested for their differentiation potential via calcium switch protocol [bib_ref] Calcium regulation of keratinocyte differentiation, Bikle [/bib_ref]. Various differentiation markers were checked via qPCR. The batch of cells showing proper differentiation and morphology were then selected for further experiments.
Proliferating keratinocytes were maintained in low Ca 2+ E-media (0.05 mM). For differentiation, they were allowed to reach 100% confluence and then introduced with high Ca 2+ (1.2 mM) E-media for 48 hours. Once they differentiated and appear as sheet-like morphology, scratch wounds were made (with the help of a 1-ml tip) at multiple sites in each culture plate. To keep the constancy between experiments, the distance between the consecutive scratch was kept approximately 0.5 mm. The scratch wounds were followed by a 1× PBS wash, and fresh high Ca 2+ (1.2 mM) E-media were added to each plate. As described in the figure legends, the cells were harvested at several time points using TRIzol reagent for RNA isolation or using lysis buffer for DNA isolation.
## Mice
C57Bl6/J animals were originally purchased from Jackson Laboratory (Stock No. 000664) and were bred for >10 generations in the NCBS vivarium facility. The 8-week-old mice were anesthetized, and 5-mm or 8-mm punch biopsies were used to make full-thickness excisional wounds.
## Tissue section and staining
Wounded regions were embedded in OCT, frozen on dry ice, and stored in −80 o freezer for further sectioning and antibody staining, and 10-to 15-μm section were taken, stained with primary antibody at 4˚C overnight, and then with secondary antibody at RT for 20 to 30 minutes. Antibodies used in this study are as following: Caspase-8 (Enzo #ALX-804-447-C100), DNMT3a (Abcam #ab2850, SC #365769), K5 (lab generated). Sections were imaged using IX73 Olympus microscope.
In situ hybridization DIG labeled 5 0 mouse caspase-8 cRNA probe was synthesized as per the manufacturer's instructions (Roche dig labeling kit-# 11175025910). In situ hybridization was performed as described earlier [bib_ref] In situ hybridization on mouse paraffin sections using DIG-labeled RNA probes, Wu [/bib_ref]. Briefly, the paraffin tissue sections were deparaffinized by treatment by xylene and ethanol gradient, or the 4% PFA fixed cells were permeabilized using 0.2% Tri-tonX-100 for 10 minutes at room temperature, and 5 ng DIG labeled cRNA probes per 100 μL hybridization buffer was applied on the sections overnight at 63˚C. Same concentration of DIG labeled mRNA with the complimentary sequence to cRNA was used as a negative control. Washing was done at 65˚C. The Anti-DIG antibody (Roche # 11093274910 Roche) was applied overnight as per manufacturer's instructions. Sections were developed for 30 minutes at 37˚C using BCIP/NBT solution (Sigma # B6404). Reaction was stopped using de-ionized water once the purple color was developed. Sections were mounted using MOWIOL solution and imaged using bright-field microscope.
## Hydrogel of varying stiffness
The polyacrylamide-based hydrogels were prepared as describe in [bib_ref] Preparation of hydrogel substrates with tunable mechanical properties, Tse [/bib_ref] [bib_ref] Simple polyacrylamide-based multiwell stiffness assay for the study of stiffness-dependent cell responses, Syed [/bib_ref]. They were coated with collagen and seeded with enough cells to make it 80% to 100% confluent and were allowed to settle for 24 to 48 hours before initiating the keratinocyte differentiation.
## Chromatin immunoprecipitation (chip)
ChIP of histone modification was performed as described previously [bib_ref] Analysis of epigenetic modifications of chromatin at specific gene loci by native..., Brand [/bib_ref] with some modifications. In brief, harvested keratinocytes (unscratched and scratched) were cross-linked with 1% formaldehyde. Cells were lysed in buffer N containing DTT, PMSF, and 0.3% NP-40. After isolation of nuclei, chromatin fractionation was done using 0.4 U of MNase (N5386, Sigma) at 37˚C for 10 minutes. Reaction was stopped using MNase stop buffer without proteinase K. Simultaneously, antibodies against H3K27me3, H3K9me3, H3K4me3, H3K9ac, and Rabbit IgG were kept for binding with Dynabeads for 2 hours at RT. After equilibration of beads, chromatin was added for pre-clearing. To antibody bound beads, pre-cleared chromatin was added and kept for IP at 4˚C overnight. Next day, beads were washed and eluted at 65˚C for 5 minutes. Eluted product was subjected to reverse cross-linking along with input samples, first with RNase A at 65˚C overnight and then with Proteinase K at 42˚C for 2 hours. After reverse cross-linking, DNA purification was performed using phenol:chloroform extraction method. Antibodies used for this protocol are listed here: H3K27me3 (07-449, Milipore), H3K9me3 (ab8898, Abcam), H3K9ac (ab4441, Abcam), H3K4me3 (ab8580, Abcam).
## Bisulphite reaction, sequencing, and analysis
Genomic DNA was isolated by salting out method as described elsewhere [bib_ref] A simple salting out procedure for extracting DNA from human nucleated cells, Miller [/bib_ref] , then treated with RNase for 1 hour at 37˚C. Further, approximately 20 μg DNA was taken in 200 μL volume and purified with phenol:chloroform extraction method. The purified DNA was checked for its integrity via running on the agarose gel. The DNA sample having good integrity and free of RNA were taken for bisulphite conversion as per manufacturer's protocol (Zymo #D5005). The converted DNA was then amplified using bisulphite conversion-specific primers, the amplified product was assessed on the agarose gel, ligated with TOPO-TA vector, and transformed in competent Top10 cells. Two to 3 colonies from each experiment were sent for Sanger's sequencing using caspase-8 promoter-specific sequencing primers (GAATAAGGAAGT GTTTTTTAG, AAAACTATACTCACTTCCTATTC). The sequenced file (FASTA) was uploaded to http://quma.cdb.riken.jp/ for CpG methylation analysis.
## Lentivirus shrna constructs and transduction
Plasmids expressing shRNAs were obtained from TransOmics (DNMT3a # TLMSU1400). To produce viruses, HEK293T cells were transfected with psPAX2, pMD2.G, and either non-targeting random RNA sequence vector or shRNA-containing plasmids, using Lipofectamine transfection reagent according to the manufacturer's protocol. Following a 48-to 72-hour transfection, the virus particle-containing media was collected, concentrated with filters, and added to the differentiated cells for 24 hours. Expression of DNMT3a was measured 2 to 3 days after viral infection. Silencing efficiency was confirmed by immunoblotting.
## Quantitative real-time pcr
RNA was isolated from human keratinocytes (proliferating or differentiated) using the RNAiso Plus (Takara), and 1 μg of RNA was used to prepare cDNA using the PrimeScript kit (Takara). cDNA equivalent to 100 ng of RNA was used for setting up the qPCR reaction using the SYBR green 2× master mix. All reactions were performed in technical triplicates using the CFX384 Touch Real-Time PCR detection system (BioRad). Primers used in this study are listed here: caspase-8 mRNA (TCTGCTGGGAATGGCTACGGTGAA, GTGTGAAGGTGGGCTGTGG CATCT), caspase-8 promoter (GGGAATAAGGAAGTGTCCTCCA, CCCAGAACTGTACT CACTTCCTG), beta actin (GGGCTATGCTCTCCCTCAC, GATGTCACGCACGATTTCC).
## Rna sequencing and data analysis
The scratch wounded cells and controls were collected after 8 hours in TRIzol reagent and RNA was isolated using standard TRIzol-based RNA isolation method. The library preparation and NGS RNA sequencing steps were outsourced to a commercial facility (Genotypic). Once the raw sequencing reads were received, sequencing data analysis was performed using the following analysis pipeline. Briefly, raw sequencing data was QC checked with the "FASTQC" tool (Babraham Bioinformatics). Adapter contamination and bad quality reads were trimmed using "Trimmomatic" tool [bib_ref] Trimmomatic: A flexible trimmer for Illumina sequence data, Bolger [/bib_ref]. The good quality reads were then mapped to mm10 (mouse) reference genome using "HISAT2" [bib_ref] HISAT: A fast spliced aligner with low memory requirements, Kim [/bib_ref]. The resulting "SAM" outputs were converted to "BAM" output and sorted. The "HTSeq-Count" tool was used to generate expression matrix from all 4 samples. Then, differential expression was analyzed with the help of DESeq2 R package.
## Gene ontology enrichment analysis
To explore enrichment of Gene Ontology among the significantly down-regulated (n = 428) and up-regulated genes (n = 358), we have used http://geneontology.org/ resources that runs "PANTHER" for the enrichment analysis [bib_ref] PANTHER: A library of protein families and subfamilies indexed by function, Thomas [/bib_ref]. Additional details of the NGS RNA seq samples are given in [fig_ref] Table 1: NGS read counts [/fig_ref].
## Supporting information
[fig] Fig 1: Kinetics of caspase-8 promoter methylation and expression. (A) Levels of caspase-8 mRNA at different time points post-scratch wound (fold change) (n = 4). (B) In vitro ISH of caspase-8 mRNA showing its levels at scratch margins over time [scale = 10 μm]. (C) In vivo ISH of caspase-8 mRNA showing its levels at wound proximal and distal regions over time (dotted line represents basement membrane, Epi = Epidermis, Der = Dermis) [scale = 20 μm]. (D) Bisulphite sequencing of caspase-8 promoter proximal region (265 bp) shows methylation status of 10 individual CpG sites (columns) from 10 cloned PCR products (rows) at various time points post-scratch wound. Percentage value denotes the percent methylation for each group of CpG sites over time (refer S1D Fig for the sequenced region and primer sites, n = 5 with 2 technical replicates). (Data cytosolic localization of DNMT3a in differentiated primary epidermal keratinocytes (S2C Fig). [/fig]
[fig] Fig 2: De novo DNMT3a increases nuclear localization at wound proximal site. (A) DNMT3a and DAPI staining of wound proximal (<0.5 mm) skin sections at different time interval (small white arrows showing nuclei of suprabasal keratinocytes, negative [/fig]
[fig] Fig 4: Effect of cellular tension on DNMT3a localization and caspase-8 expression. (A) Effect of EGTA and blebbistatin on the localization of DNMT3a. (B) Effect of scratch wound DNMT3a localization in presence and absence of calyculin-A. (C) Effect of various matrix stiffness on the localization of DNMT3a. (D) Fold change in the levels of caspase-8 mRNA as a result of varios pharmacological and mechanical approaches of tension modulation (n = 4), [scale bar = 20 μm]. (Data are shown as mean ± SEM, P-values were calculated [/fig]
[fig] S1: Fig. Caspase-8 RNA half-life and CpG positions on its promoter proximal region. (A) Quantification of caspase-8 mRNA to check its half-life post transcriptional block (using Actinomycin-D) (n = 3). (B) In situ hybridization with anti-sense and sense probe of caspase-8 RNA (in vitro) [scale = 10 μm]. (C) In situ hybridization with anti-sense and sense probe of caspase-8 RNA (in vivo) [scale = 20 μm]. (D) Model showing positions of CpG dinucleotide and SP1 binding sites in caspase-8 promoter proximal region. (Data are shown as mean ± SEM, P-values were calculated using 1-way ANOVA with Dunnett's test (A), � P � 0.05, �� P � 0.01, ��� P � 0.001, ns = P > 0.05.) Data underlying the graphs can be found in A Fig of Raw Data. (TIF) S2 Fig. (A) Representative image of unwounded/wound-distal skin section stained with DNMT3a, DAPI, and K5. (B') A model showing the quantification method of DAPI and DNMT3a stain intensities over the line of interest (1, 2) from proliferating and differentiated keratinocytes, followed by (B") the plots of intensity values (gray unit) (calculated intensities from 4 biological replicates). Staining of in vitro proliferating and differentiated keratinocytes with (C), DNMT3a/DAPI and (D), DNMT3b/DAPI. (E) DNMT3b/DAPI staining of scratch wounded in vitro differentiated keratinocytes. (F) DNMT3a western blot analysis from control and scratch wounded keratinocytes at 8-hour time point (G), DNMT3a, DAPI, and K5 staining of a completely healed mouse skin section [scale = 20 μm]. Data underlying the graphs can be found in S2B Fig of Raw Data. (TIF) S3 Fig. (A) qPCR analysis of caspase-8 mRNA in scratch wounded keratinocytes, pre-treated with 5-Aza-2 0 -deoxycytidine (5A-dC) or DMSO (n = 4) (B), western blot analysis from keratinocytes transduced with scrambled RNA or DNMT3a shRNA (α-Tub = alpha-tubulin) (data are shown as mean ± SEM, P-values were calculated using 2-tailed t test (A), � P � 0.05, �� P � 0.01, ��� P � 0.001, ns = P > 0.05). Data underlying the graphs can be found in S3A Fig and S3B Fig of Raw Data. (TIF) S4 Fig. (A) Model showing various potential protein molecules (red labels) involved in generating and/or sensing the cellular tension. (B) Quantification of DNMT3a localization (nuclear v/s cytoplasmic) in EGTA and blebbistatin-treated keratinocytes compared to control (n = 3). (C) Quantification of DNMT3a localization in scratch wound proximal (�100 μm) keratinocytes comparing control and calyculin-A-treated scratch wounds (n = 3). (D) Quantification of DNMT3a localization in (n = 3). (Data are shown as mean ± SEM, P-values were calculated using 2-tailed t test (D), � P � 0.05, �� P � 0.01, ��� P � 0.001, ns = P > 0.05.) Data underlying the graphs can be found in S4B-S4D Fig of Raw Data. (TIF) S5 Fig. Methylation-induced transcriptional reprogramming of epidermal keratinocytes from homeostasis to repair. (A) Heat map of differentially regulated genes in control and scratch wounded keratinocytes. (B) Scratch wound induced transcriptional down-regulation of genes and status of their associated DNA methylation levels. (C) Fold change of transcriptionally up-regulated genes and their associated DNA methylation levels (MeDIP-qPCR, yaxis = fold change compared to control). (D) DNMT3a and caspase-8 staining of control and psoriatic mouse skin (induced through imiquimod treatment), [scale bar = 100 μm]. Data underlying the graphs can be found in S5A-S5C Fig of Raw Data. DNMT3a, DNA methyltransferase 3A; MeDIP, Methylated DNA Immunoprecipitation. (TIF) S6 Fig. Gene ontology of up-regulated and down-regulated genes (Biological Processes). Processes are listed as-Log 10 of adjusted FDR values. Top 15 relevant biological processes are chosen for generating the graphs. Data underlying the graphs can be found in S6 Fig of Raw Data. (TIF) Raw Data. Raw data for Figs 1-4 and -S6. (ZIP) [/fig]
[table] Table 1: NGS read counts. https://doi.org/10.1371/journal.pbio.3001777.t001 [/table]
|
Dichorionic quadruplet pregnancy comprising monozygotic triplets and singleton after intracytoplasmic sperm injection and transfer of two fresh embryos: a case report
Monozygotic triplet pregnancies are very rare in assisted reproductive technology, and the relationship between monozygotic multiple pregnancies and several assisted reproductive techniques, including blastocyst transfer, remains unclear. Here, the case of a 28-year-old female patient with dichorionic quadruplet pregnancy following intracytoplasmic sperm injection and transfer of two day-3 fresh embryos, without assisted hatching, is reported. At 7 weeks following embryo transfer, the dichorionic quadruplet pregnancy, comprising monozygotic monochorionic triamniotic (MCTA) triplets plus a singleton, was detected by a transabdominal ultrasound scan. After counselling, the patient underwent selective reduction of the MCTA triplet pregnancy at 7 weeks after embryo transfer. The remaining singleton pregnancy was uneventful, resulting in a live birth at 38 þ weeks. As the predictors of monozygotic multiple gestations remain poorly characterized, clinicians and patients should give great consideration to the risks associated with monozygotic multiple pregnancies, even if the patient has not undergone blastocyst transfer.
# Introduction
Multiple pregnancies following assisted reproductive technology (ART) have significantly increased over recent decades. According to statistics reports from the US Centres for Disease Control and Prevention National ART Surveillance System, 1 ART-conceived multiple-birth infants contributed to [bib_ref] Monozygotic triplet pregnancies after single blastocyst transfer: two cases and literature review, Dessolle [/bib_ref].75% of all multiple-birth infants (19 570 of 132 703) in 2017 and 14.67% of all twins (18 890 of 128 774), and 17.3% of all triplets and higher-order infants (680 of 3929) were accounted for by ART. [bib_ref] Assisted reproductive technology surveillance -United States, Sunderam [/bib_ref] However, multiple pregnancies, particularly monozygotic multiple pregnancies, are often at higher maternal and fetal risks, such as twin to twin transfusion syndrome (TTTS) and selective intrauterine growth restriction. [bib_ref] Determinants of monozygotic twinning in ART: a systematic review and a meta-analysis, Hviid [/bib_ref] Therefore, single embryo transfer is highly recommended and currently considered to be the main method for minimizing the risk of multiple pregnancies associated with in vitro fertilization (IVF). Although the rate of single embryo transfers was significantly higher in 2017 than in 2016, 1,3 clinicians still need to pay attention to monozygotic multiple pregnancies, as a special type of multiple pregnancy with high risk and potentially poor outcome.
In order to achieve improvements in pregnancy rates while decreasing the rate of multiple pregnancies, prolonged culture and single blastocyst transfer has become more common. It is important to note, however, that single embryo transfer in ART cannot completely prevent the occurrence of multiple pregnancy. [bib_ref] Determinants of monozygotic twinning in ART: a systematic review and a meta-analysis, Hviid [/bib_ref] ART itself increases the incidence of multiple births, and monozygotic multiple pregnancies occur more frequently in ART gestations than in spontaneous pregnancies. [bib_ref] High incidence of monozygotic twinning in infertility treatment, Sobek [/bib_ref] The prevalence in Japan of multiple pregnancy with zygotic splitting was reported to be 1.36% of ART pregnancies between 2007 and 2014. [bib_ref] Prevalence and risk factors of zygotic splitting after 937 848 single embryo..., Ikemoto [/bib_ref] The mechanism of spontaneous zygotic splitting remains unknown, but many factors, including maternal age, culture media, ovulatory induction, cryopreservation, blastocyst transfer, prolonged culture and micro-manipulation of the zona pellucida, such as intracytoplasmic sperm injection (ICSI) and assisted hatching, are thought to be associated with monochorionic multiple pregnancy. 6,7 Compared with monozygotic twinning pregnancies, the occurrence of monozygotic triplet pregnancies is very rare in ART, but the risks and complications of monozygotic triplet pregnancy are significantly increased compared with monozygotic twin pregnancies. [bib_ref] Monochorionic triplet gestation after in vitro fertilization using donor oocytes: case report..., Henne [/bib_ref] Monozygotic high-order pregnancies require increased attention in assisted reproduction. Herein, a case of dichorionic quadruplet pregnancy, comprising monozygotic triplets and a singleton, after intracytoplasmic sperm injection and transfer of two fresh embryos without assisted hatching, is presented, together with a specific review of possible risk factors for the occurrence of monozygotic high-order pregnancies in ART. the patient provided written informed consent for publication of the case. The reporting of this study conforms to CARE guidelines. [bib_ref] The CARE guidelines: consensus-based clinical case reporting guideline development, Gagnier [/bib_ref] In August 2016, a 28-year-old female patient (G2P1, with no obstetric-related comorbidities) and her husband received a second IVF cycle with ICSI at the Reproductive Centre, West China Second Hospital, Sichuan University, China, due to obstructive azoospermia. The female patient had a history of regular menstrual cycle; and physical examinations, body mass index, hormonal status and family history of multiple pregnancies were unremarkable for both participants. They had received their first ICSI cycle in 2011, while the female patient was aged 23 years, and obtained four embryos. Transfer of two frozen-thawed embryos in June 2011 resulted in a healthy male infant by vaginal delivery at 38 weeks' gestation in April 2012. Transfer of the remaining two frozen-thawed embryos in 2016 failed to achieve successful pregnancy, so the couple underwent a second cycle of ICSI in 2016.
Controlled ovarian hyperstimulation was performed using a gonadotrophinreleasing hormone agonist long protocol, in mid-luteal phase. Pituitary downregulation was initiated on day 21 of the previous cycle by administration of 0.1 mg triptorelin, subcutaneous injection, every other day. Ovarian stimulation was performed with daily intramuscular injections of 225 IU highly-purified urinary follicle stimulating hormone after pituitary down regulation. The total dose of gonadotropins was 2850 IU. After 10 days of gonadotrophin stimulation, human chorionic gonadotrophin (hCG) was triggered with 10 000 IU hCG when at least two follicles were >18 mm in diameter, followed by transvaginal ultrasound-guided aspiration 36 h later. A total of 13 oocytes were retrieved, 10 of which were in metaphase II and were microinjected with spermatozoa obtained by testicular sperm aspiration. On day 3 of incubation, two fresh embryos (compact and 8-cell stage) were transferred without performing assisted hatching, and the remaining three embryos were cryopreserved. The patient received progesterone supplementation with 90 mg progesterone gel, vaginally, daily. Her serum b-hCG level was found to be 144 mIU/ml at 10 days following embryo transfer, and transabdominal ultrasound, performed at 35 days, showed two intrauterine gestational sacs with detectable heart beats [fig_ref] Figure 1: Representative image from a transabdominal ultrasound scan performed at 35 days after... [/fig_ref]. Transabdominal ultrasound, repeated at 7 weeks after embryo transfer, revealed three embryonic buds in one of the gestational sacs. Subsequent transvaginal ultrasound confirmed an intrauterine quadruplet pregnancy, consisting of monozygotic monochorionic triamniotic (MCTA) triplets, and a monozygotic monochorionic monoamniotic (MCMA) singleton, with a detectable heartbeat in all four fetuses. The diameters of the triplet fetuses were 1.9 cm, 2.0 cm and 2.1 cm, respectively, The couple was informed about the increased maternal and fetal risks of higher order monozygotic gestations and they were counselled regarding the possibility of a fetal reduction procedure. After extensive counselling, and considering that they had a healthy male child, they decided to undergo selective embryo reduction, and only proceed with the singleton pregnancy. Thus, at 7 weeks after embryo transfer, selective reduction was performed by ultrasound-guided transvaginal aspiration targeted at the MCTA triplets. This resulted in the cessation of cardiac activity in the monozygotic MCTA triplets, whereas normal cardiac activity was observed in the remaining monozygotic MCMA singleton. Subsequent ultrasound examinations after 24 h confirmed the presence of an ongoing singleton pregnancy. At week 10 of gestation after embryo transfer, the remaining singleton had normal nuchal translucency (1.5 mm), and continued to develop normally following selective reduction of the triplet sac. The remainder of the pregnancy was uneventful, and resulted in vaginal delivery of a healthy male infant, weighting 3 200 g, at 38 weeks and 6 days of gestation.
# Discussion
This case report describes a quadruplet pregnancy comprising one implanted embryo that developed into a monozygotic monochorionic embryo and the other that split into monozygotic MCTA embryos after the transfer of two fresh embryos generated by ICSI. This case adds to the small number of monozygotic higher-order pregnancies in ART that currently exist in the literature.
The incidence of triplet births is very low, and was reported to be 0.01% of births in the Netherlands in 1980. 10 Monozygotic triplet pregnancies are even more rare, estimated to occur in only 0.004% of natural pregnancies, and found to account for 10% of all triplet pregnancies in a population-based study. Following the introduction and development of ART, the rate of triplet births has increased significantly. The occurrence of triplets and higher-order multiples is estimated to be between 0.1% and 0.2% of pregnancies in the USA, with ART being more commonly associated with triplet pregnancies than twin or singleton pregnancies. [bib_ref] Assisted reproductive technology surveillance -United States, Sunderam [/bib_ref] [bib_ref] Assisted reproductive technology surveillance -United States, Sunderam [/bib_ref] [bib_ref] Assisted reproductive technology surveillance -United States, Sunderam [/bib_ref] In the USA, ART-conceived triplets and higher-order infants contributed to 33.0% of all triplets and higher-order infants in 2010, [bib_ref] Assisted reproductive technology surveillance -United States, Sunderam [/bib_ref] however, with the increasing use of single embryo transfer, the rate decreased year on year to 17.3% in 2017. Similar to the USA, one-third of triplet pregnancies in the Netherlands are due to ART. 10 A population-based study in 2016 suggested a 60% increase in monozygotic twinning in ART gestations versus natural pregnancy. [bib_ref] Risk of monozygotic twins after assisted reproduction: a population-based approach, Parazzini [/bib_ref] A study published in 2018 reported a prevalence of 1.36% for multiple pregnancy resulting from zygotic splitting in Japan, and the prevalence of triplets in ART pregnancies was 0.04%. [bib_ref] Prevalence and risk factors of zygotic splitting after 937 848 single embryo..., Ikemoto [/bib_ref] Millions of IVF babies have been born since 1978, and in the largest study to date, Yamashita et al., [bib_ref] Analysis of 122 triplet and one quadruplet pregnancies after single embryo transfer..., Yamashita [/bib_ref] reported 122 triplet pregnancies and one quadruplet pregnancy after single embryo transfer in Japan between 2007 and 2014. Apart from the study by Yamashita et al., [bib_ref] Analysis of 122 triplet and one quadruplet pregnancies after single embryo transfer..., Yamashita [/bib_ref] only just over 30 cases of monozygotic triplet pregnancy in ART have been reported to date worldwide. Available published data are based on limited populationbased sample studies, small sample studies or case reports. [bib_ref] Live birth of monochorionic triamniotic triplets after in vitro fertilization and blastocyst..., Radwan [/bib_ref] [bib_ref] Monochorionic triplet gestation after in vitro fertilization using donor oocytes: case report..., Henne [/bib_ref] The occurrence of monozygotic triple and high-order pregnancies is thought to increase in ART due to similar split mechanisms associated with monozygotic twinning; several procedural factors in ART may be associated with the mechanisms, but the specific ART procedures that lead to splitting remain unknown. [bib_ref] Determinants of monozygotic twinning in ART: a systematic review and a meta-analysis, Hviid [/bib_ref] [bib_ref] High incidence of monozygotic twinning in infertility treatment, Sobek [/bib_ref] [bib_ref] Analysis of 122 triplet and one quadruplet pregnancies after single embryo transfer..., Yamashita [/bib_ref] [bib_ref] Risk of monozygotic twins after assisted reproduction: a population-based approach, Parazzini [/bib_ref] [bib_ref] Monochorionic-triamniotic triplet pregnancy after intracytoplasmic sperm injection, assisted hatching, and twoembryo transfer:..., Ghulmiyyah [/bib_ref] [bib_ref] Monozygotic twins and triplets in association with blastocyst transfer, Jain [/bib_ref] [bib_ref] Monozygotic triplets after single blastocyst transfer: case report and literature review, Lee [/bib_ref] [bib_ref] Monochorionic triplet and monoamniotic twins gestation after intracytoplasmic sperm injection and laser-assisted..., Pantos [/bib_ref] [bib_ref] Monochorionic triplets after single embryo transfer, Risquez [/bib_ref] [bib_ref] Monozygotic multiple gestation following in vitro fertilization: analysis of seven cases from..., Yanaihara [/bib_ref] [bib_ref] Increased monozygotic twinning rate after ovulation induction, Derom [/bib_ref] [bib_ref] Blastocyst culture is associated with an elevated incidence of monozygotic twinning after..., Kawachiya [/bib_ref] [bib_ref] Prevalence and risk factors of monochorionic diamniotic twinning after assisted reproduction: A..., Song [/bib_ref] In the present study, the limited published reports regarding ART-conceived monozygotic triplet and higher-order pregnancies were reviewed to analyse possible factors.
Several factors relating to ART procedures are thought to be associated with the occurrence of monozygotic triple and higher-order pregnancies. Identifying which specific ART procedures have led to the occurrence of monozygotic triple pregnancies remains difficult due to limited relevant reports in the literature. Many studies have implied a correlation between blastocyst transfer and monozygotic twinning, and suggest that blastocyst transfer is a risk factor for monozygotic twinning. [bib_ref] Live birth of monochorionic triamniotic triplets after in vitro fertilization and blastocyst..., Radwan [/bib_ref] [bib_ref] Analysis of 122 triplet and one quadruplet pregnancies after single embryo transfer..., Yamashita [/bib_ref] [bib_ref] Monochorionic triamniotic triplets following conventional in vitro fertilization and blastocyst transfer, Gurunath [/bib_ref] [bib_ref] Monozygotic twins and triplets in association with blastocyst transfer, Jain [/bib_ref] [bib_ref] Monozygotic triplets after single blastocyst transfer: case report and literature review, Lee [/bib_ref] [bib_ref] Monozygotic triplets and monozygotic twins after ICSI and transfer of two blastocysts:..., Unger [/bib_ref] [bib_ref] Monozygotic multiple gestation following in vitro fertilization: analysis of seven cases from..., Yanaihara [/bib_ref] [bib_ref] Blastocyst culture is associated with an elevated incidence of monozygotic twinning after..., Kawachiya [/bib_ref] [bib_ref] Prevalence and risk factors of monochorionic diamniotic twinning after assisted reproduction: A..., Song [/bib_ref] The incidence of monozygotic twinning in ART with blastocyst transfers has increased compared with day 2-3 transfers. [bib_ref] Determinants of monozygotic twinning in ART: a systematic review and a meta-analysis, Hviid [/bib_ref] Due to the similar split mechanism with monozygotic twinning, the incidence of monozygotic triple and high-order pregnancies is thought to be increased in ART. 10,40 From reports on monozygotic higher-order pregnancies published to date, it appears that 56.2% (18/32) have occurred after blastocyst transfer, 9.4% (3/32) after day 4 embryo transfer, and 34.4% (11/32) after day 3 embryo transfer [fig_ref] Table 1: Summary of the present case and previously published reports of monozygotic triple/quadruplet... [/fig_ref]. [bib_ref] Live birth of monochorionic triamniotic triplets after in vitro fertilization and blastocyst..., Radwan [/bib_ref] [bib_ref] Monochorionic triplet gestation after in vitro fertilization using donor oocytes: case report..., Henne [/bib_ref] [bib_ref] Monozygotic triplet pregnancy following transfer of frozen-thawed embryos, Belaisch-Allart [/bib_ref] [bib_ref] Monozygotic triplet pregnancies after single blastocyst transfer: two cases and literature review, Dessolle [/bib_ref] [bib_ref] Monozygotic triplet pregnancy following egg donation and transfer of single frozenthawed embryo, Faraj [/bib_ref] [bib_ref] Dichorionic twins and monochorionic triplets after the transfer of two blastocysts, Ferreira [/bib_ref] [bib_ref] Monochorionic-triamniotic triplet pregnancy after intracytoplasmic sperm injection, assisted hatching, and twoembryo transfer:..., Ghulmiyyah [/bib_ref] [bib_ref] Monochorionic triamniotic triplets following conventional in vitro fertilization and blastocyst transfer, Gurunath [/bib_ref] [bib_ref] Monozygotic multiple gestation after intracytoplasmic sperm injection and preimplantation genetic diagnosis, Haimov-Kochman [/bib_ref] [bib_ref] Monochorionic triamniotic triplet pregnancies with assisted reproductive technology: two case reports, Iwamoto [/bib_ref] [bib_ref] Monozygotic twins and triplets in association with blastocyst transfer, Jain [/bib_ref] [bib_ref] Monozygotic triplets after single blastocyst transfer: case report and literature review, Lee [/bib_ref] [bib_ref] Dichorionic quadramniotic quadruple gestation with monochorionic triamniotic triplets after two embryos transfer..., Li [/bib_ref] [bib_ref] Monozygotic quadruplets after in vitro fertilization and embryo transfer, Liu [/bib_ref] [bib_ref] Monochorionic triplet and monoamniotic twins gestation after intracytoplasmic sperm injection and laser-assisted..., Pantos [/bib_ref] [bib_ref] Monochorionic triplets after single embryo transfer, Risquez [/bib_ref] [bib_ref] A case of triple monoamniotic pregnancy combined with a bioamniotic twinning after..., Salat-Baroux [/bib_ref] [bib_ref] Monochorionic quadramniotic and triamniotic pregnancies following single embryo transfers: two case reports..., Saravelos [/bib_ref] [bib_ref] Successful quintuplet pregnancy of monochorionic male quadruplets and single female after double..., Schlueter [/bib_ref] [bib_ref] Monozygotic triplets and dizygotic twins following transfer of three poor-quality cleavage stage..., Tal [/bib_ref] [bib_ref] Conjoined twins in a monochorionic triplet pregnancy after in vitro fertilization: a..., Talebian [/bib_ref] [bib_ref] Monochorionic triplets following intracytoplasmic sperm injection: a report of two consecutive cases, Ulug [/bib_ref] [bib_ref] Monozygotic triplets and monozygotic twins after ICSI and transfer of two blastocysts:..., Unger [/bib_ref] [bib_ref] Monozygotic multiple gestation following in vitro fertilization: analysis of seven cases from..., Yanaihara [/bib_ref] [bib_ref] Quintuplet pregnancy following transfer of two blastocysts: case report, Zikopoulos [/bib_ref] [bib_ref] Successful monozygotic triplet pregnancy after a single blastocyst transfer following in vitro..., Ota [/bib_ref] [bib_ref] Three blastocyst stage embryo transfer resulting in a quintuplet pregnancy, Yakin [/bib_ref] Nearly one third of cases of monozygotic triple and higher-order pregnancies with ART involved transfer of day 3 cleavage stage embryos, while higher rates were associated with day 4 embryo and day 5 blastocyst transfers combined. With the increasing popularity of blastocyst transfer in many IVF centres, the monozygotic triple and higher-order pregnancies related to day 3 embryo transfer seems to have become relatively uncommon. The present study describes the third reported case in the last decade of monozygotic triple pregnancies and live birth after day 3 embryo-transfer. With current data and limited case reports, it remains difficult to draw definite conclusions, but the available data suggest that prolonged culture and blastocyst transfer may be one of the important factors in monozygotic higher-order pregnancies.
Micro-manipulation of the zona pellucida, such as ICSI, assisted hatching and biopsy, is probably another risk factor for the occurrence of monozygotic multiple pregnancies. Since the first report of a correlation between zona pellucida structure following ART and monozygotic twinning, 42 many studies have analysed the association between micro-manipulation techniques and multiple monozygotic twinning, and have found that manipulation of the zona pellucida may cause disruption and splitting of the inner cell mass and increase the rate of monozygotic twinning. [bib_ref] Determinants of monozygotic twinning in ART: a systematic review and a meta-analysis, Hviid [/bib_ref] [bib_ref] Monochorionic-triamniotic triplet pregnancy after intracytoplasmic sperm injection, assisted hatching, and twoembryo transfer:..., Ghulmiyyah [/bib_ref] [bib_ref] Monozygotic multiple gestation after intracytoplasmic sperm injection and preimplantation genetic diagnosis, Haimov-Kochman [/bib_ref] [bib_ref] Monochorionic triplet and monoamniotic twins gestation after intracytoplasmic sperm injection and laser-assisted..., Pantos [/bib_ref] [bib_ref] Monozygotic triplets and monozygotic twins after ICSI and transfer of two blastocysts:..., Unger [/bib_ref] [bib_ref] Identical twins and in vitro fertilization, Edwards [/bib_ref] An increased rate of monozygotic twinning has been shown after ICSI and assisted hatching, and the largest study to analyse triplet or quadruplet pregnancies after single embryo transfer reported that blastocyst cultures and assisted hatching (P ¼ 0.002 and P < 0.001, respectively) are risk factors for monozygotic twinning. [bib_ref] Analysis of 122 triplet and one quadruplet pregnancies after single embryo transfer..., Yamashita [/bib_ref] However, some studies do not support any association between a higher monozygotic twinning rate and micro-manipulation of the zona pellucida. [bib_ref] Determinants of monozygotic twinning in ART: a systematic review and a meta-analysis, Hviid [/bib_ref] [bib_ref] High incidence of monozygotic twinning in infertility treatment, Sobek [/bib_ref] [bib_ref] Risk of monozygotic twins after assisted reproduction: a population-based approach, Parazzini [/bib_ref] [bib_ref] Prevalence and risk factors of monochorionic diamniotic twinning after assisted reproduction: A..., Song [/bib_ref] [bib_ref] Monozygotic twinning after assisted reproductive technologies: a case report of asymmetric development..., Tocino [/bib_ref] Based on the published case reports summarised in [fig_ref] Table 1: Summary of the present case and previously published reports of monozygotic triple/quadruplet... [/fig_ref] In addition, a retrospective observational study in Japan, involving 937 848 single embryo transfer cycles, showed that embryo manipulations using blastocyst transfer, assisted hatching and frozen-warmed embryo transfer were potential risk factors for zygotic splitting, however, ICSI was not a potential risk factor. [bib_ref] Prevalence and risk factors of zygotic splitting after 937 848 single embryo..., Ikemoto [/bib_ref] However, in the present case, the transfer was performed with day 3 fresh embryos without assisted hatching, and it appears that ICSI may have been the possible risk factor for the monozygotic multiple pregnancies. Nonetheless, the relationship between ICSI and monozygotic multiple pregnancies remains controversial, and further studies are required to clarify this association. The correlation between maternal age or oocyte age and monochorionic multiple pregnancy has been analysed in the studies summarised in [fig_ref] Table 1: Summary of the present case and previously published reports of monozygotic triple/quadruplet... [/fig_ref]. The mean age associated with monochorionic high-order pregnancy was found to be 31.2 AE 4.5 years after excluding four cases (three of which used donor oocytes and one that did not report the age). Twenty-nine of the cases used their own eggs, five patients were older than 35 years, and 24 patients were aged less than 35 years. A metaanalysis showed that younger maternal age may be associated with monozygotic twinning, 2 however, the largest study found no difference in age between singleton pregnancies and monozygotic triplet or quadruplet pregnancies. 7 Therefore, the potential association between age and monozygotic multiple pregnancies requires further investigation. In all cases, only seven out of 32 patients received frozen/ thawed embryo transfer; thus, embryonic freezing does not appear to play an important role in the incidence of monozygotic multiple pregnancies, which is supported by a previously published study. [bib_ref] Analysis of 122 triplet and one quadruplet pregnancies after single embryo transfer..., Yamashita [/bib_ref] Because of the high risks and particular complications associated with monozygotic multiple pregnancies, selective fetal reduction to twins or singleton is an option to improve perinatal outcome. Nearly all of the couples described in the previously published cases were stated to have been informed of the increased maternal and fetal risks of monozygotic higher order gestations, and were counselled regarding the possibility of a fetal reduction procedure. From the results of limited reports summarised in [fig_ref] Table 1: Summary of the present case and previously published reports of monozygotic triple/quadruplet... [/fig_ref] , 43.7% of patients (14/32) underwent the fetal reduction procedure, and most cases resulted in a successive pregnancy or live birth. In cases of monozygotic triplet pregnancies combined with another singleton or twin pregnancy, the remaining pregnancy was often reported to have better outcomes after reduction of the MCTA. [bib_ref] Monozygotic multiple gestation after intracytoplasmic sperm injection and preimplantation genetic diagnosis, Haimov-Kochman [/bib_ref] [bib_ref] Monochorionic triplet and monoamniotic twins gestation after intracytoplasmic sperm injection and laser-assisted..., Pantos [/bib_ref] [bib_ref] Monozygotic triplets and dizygotic twins following transfer of three poor-quality cleavage stage..., Tal [/bib_ref] [bib_ref] Quintuplet pregnancy following transfer of two blastocysts: case report, Zikopoulos [/bib_ref] [bib_ref] Three blastocyst stage embryo transfer resulting in a quintuplet pregnancy, Yakin [/bib_ref] Reducing one or two fetuses in monozygotic triplet pregnancies may lead to subsequent MCTA fetal death, and some cases with no fetal reduction surgery resulted in live births. [bib_ref] Monozygotic triplet pregnancies after single blastocyst transfer: two cases and literature review, Dessolle [/bib_ref] [bib_ref] Monozygotic twins and triplets in association with blastocyst transfer, Jain [/bib_ref] [bib_ref] Dichorionic quadramniotic quadruple gestation with monochorionic triamniotic triplets after two embryos transfer..., Li [/bib_ref] [bib_ref] Monochorionic quadramniotic and triamniotic pregnancies following single embryo transfers: two case reports..., Saravelos [/bib_ref] [bib_ref] Monozygotic triplets and dizygotic twins following transfer of three poor-quality cleavage stage..., Tal [/bib_ref] [bib_ref] Monochorionic triplets following intracytoplasmic sperm injection: a report of two consecutive cases, Ulug [/bib_ref] [bib_ref] Monozygotic triplets and monozygotic twins after ICSI and transfer of two blastocysts:..., Unger [/bib_ref] Although fetal reduction may significantly reduce the maternal and neonatal risk in other twin and multiple pregnancies, it remains unclear whether selective fetal reduction benefits monozygotic triplet pregnancies.
In conclusion, in addition to blastocyst transfer, the relationship between monozygotic multiple pregnancies and several assisted reproductive techniques, such as manipulation of the zona pellucida, remains unclear. As the predictors of monozygotic multiple gestations are poorly characterized, patients should be informed of the risks of monozygotic multiple pregnancies after assisted reproductive techniques. Both patients and infertility specialists need to pay great attention to the risks associated with monozygotic multiple pregnancies, even if the patient only receives general assisted reproductive technology, and does not undergo blastocyst transfer.
[fig] Figure 1: Representative image from a transabdominal ultrasound scan performed at 35 days after embryo transfer showing two intrauterine gestational sacs. and the diameter of the singleton was 2.3 cm(Figure 2). [/fig]
[fig] Figure 2: Representative image from a transvaginal ultrasound scan performed at 7 weeks after embryo transfer showing monochorionic triamniotic triplets (white arrows; three viable embryos were detected in one gestational sac); and a single embryo in another gestational sac (black arrow). [/fig]
[table] Table 1: Summary of the present case and previously published reports of monozygotic triple/quadruplet pregnancies. [/table]
|
Accurate Encoding and Decoding by Single Cells: Amplitude Versus Frequency Modulation
Cells sense external concentrations and, via biochemical signaling, respond by regulating the expression of target proteins. Both in signaling networks and gene regulation there are two main mechanisms by which the concentration can be encoded internally: amplitude modulation (AM), where the absolute concentration of an internal signaling molecule encodes the stimulus, and frequency modulation (FM), where the period between successive bursts represents the stimulus. Although both mechanisms have been observed in biological systems, the question of when it is beneficial for cells to use either AM or FM is largely unanswered. Here, we first consider a simple model for a single receptor (or ion channel), which can either signal continuously whenever a ligand is bound, or produce a burst in signaling molecule upon receptor binding. We find that bursty signaling is more accurate than continuous signaling only for sufficiently fast dynamics. This suggests that modulation based on bursts may be more common in signaling networks than in gene regulation. We then extend our model to multiple receptors, where continuous and bursty signaling are equivalent to AM and FM respectively, finding that AM is always more accurate. This implies that the reason some cells use FM is related to factors other than accuracy, such as the ability to coordinate expression of multiple genes or to implement threshold crossing mechanisms.Author SummarySignals, and hence information, can generally be transmitted either by amplitude (AM) or frequency (FM) modulation, as used, for example, in the transmission of radio waves since the 1930s. Both types of modulation are known to play a role in biology with AM conventionally associated with signaling and gene expression, and FM used to reliably transmit electrical signals over large distances between neurons. Surprisingly, FM was recently also observed in gene regulation, making their roles less distinct than previously thought. Although the engineering advantages and disadvantages of AM and FM are well understood, the equivalent question in biological systems is still largely unsolved. Here, we propose a simple model of signaling by receptors (or ion channels) with subsequent gene regulation, thus implementing both AM and FM in different types of biological pathways. We then compare the accuracy in the production of target proteins. We find that FM can be more accurate than AM only for a single receptor with fast signaling, whereas AM is more accurate in slow gene regulation and with signaling by multiple receptors. Finally, we propose possible reasons that cells use FM despite the potential decrease in accuracy. 2 nucleus [3,10,[15][16][17], in close analogy to frequency modulation (FM). (Note that, although there is no modulation of an underlying carrier wave as in radio broadcasting [18], the AM/FM terminology is commonly used in quantitative biology[10,15].) Although several hypotheses have been put forward, the benefits and detrimental effects of either type of response remain largely unclear.There is experimental evidence that both types of modulation occur in gene regulation. For example, take the budding yeast Saccharomyces cerevisiae. Under oxidative stress the nuclear concentration of transcription factor Msn2 is proportional to the H 2 O 2 concentration, suggesting an AM mechanism(Fig. 1A,B)[10]. However, in response to a calcium stimulus, Crz1, which is normally cytoplasmic, enters the nucleus in unsynchronized bursts, regulating at least a hundred target genes(Fig. 1C) [15]. The level of stimulus affects only the frequency of bursts, not their amplitude and duration, which implies FM (Fig. 1Dand inset) [15,19]. Similarly, Msn2 and its homologue Msn4 exhibit FM under glucose limitation [10]. Bursty FM is also found in bacteria and mammals, indicating that this is a general modulation scheme across different cell types. For example, during energy-depletion stress, the bacterium Bacillus subtilis activates the alternative sigma factor σ B in discrete stochastic pulses, regulating around 150 downstream genes [20]. In addition, isoform NFAT4 in activated T-cells shows similar behavior [21].What are the relative benefits of AM and FM? One important issue is the susceptibility to noise, which affects the accuracy of sensing. For example, in broadcasting radio signals it is well known that FM is less affected by noise than AM. This is because noise mainly deteriorates the amplitude, which is where the information is stored in AM. A similar argument also favors action potentials in communicating neuronal signals over long distances [22]. In contrast, it has been hypothesized that for other cell types, such as yeast, the bursty nature of FM may introduce more noise than AM, so that AM might be preferable (Fig. 2A,B)[15]. However, two recent articles (which we discuss below) disagree with this and suggest that FM may still be more accurate [23, 24]. In addition, it is important to remember that there are often other factors than noise minimization. For example, it has been suggested that, in situations where multiple genes need to be up or down regulated, FM can provide greater coordination and reliability(Fig. 2C,D)[15, 19].Mora and Wingreen considered a model for a single receptor embedded in a cell membrane and compared the noise in the output for two signaling mechanisms: continuous (CM) and bursty (BM) modulation [23]. In CM, the receptor signals continuously whenever a ligand is bound, whereas in BM the receptor signals for a short, fixed-sized burst only upon binding of a ligand. As we explain below, for multiple receptors these mechanisms become equivalent to AM and FM, respectively. By considering integral feedback control, a common network for sensing concentration ramps and precise adaptation [25][26][27], it was found that, for fast binding and unbinding, the noise in CM can be twice that from BM, suggesting that FM leads to greater accuracy. Despite this unexpected result, there are two key points that need further clarification. First, the response was only calculated to lowest order in the small-ramp parameter, thus neglecting any time dependence of the noise. Second, the derivation solely relied on the small-noise approximation, which might work well for fast signaling, but could be inadequate for slow gene regulation.Similarly, Tostevin et al. found biologically relevant parameter regimes of promoter switching in gene regulation in which an oscillating input can produce a more constant and hence less noisy protein output level than a constant input with noise [24]. Although interesting, this model is restricted to decoding and linear pathways, and requires fine-tuning. Its general applicability remains unclear, such as whether an oscillatory input signal can be replaced by random bursts and still remain more accurate than a constant input. In fact, oscillating signals are well-known to maximize target responsiveness while bypassing desensitization from constant signals [28]. They also globally entrain with its period robust to noise [29]. Such oscillators are found in circadian clocks, segmentation clocks, cell cycle, p53 DNA repair pathways, as well as nuclear factor NF-κB, epidermal growth factor ERK, cAMP and Ca 2+ signaling [17,[30][31][32][33][34][35][36][37][38][39]. This leaves the question of the relative benefits of AM and FM (with respect to random bursts) largely unanswered. 5 AM is generally more accurate than FM, we speculate why cells may still utilize FM in certain cases of gene regulation.
Introduction Cells are exposed to changing environmental conditions and need to respond to external stimuli with high accuracy, e.g. to utilize nutrients and to avoid lethal stresses . To represent (encode) chemicals in the environment, either ligand-bound receptors trigger chemical signals or ion channels allow entry of secondary messengers. These in turn activate transcription factors (TFs), which then regulate targetprotein production (decoding). In eukaryotic cells, the conventional view is that the level of signaling within the cell directly encodes the external stimuli, with consequent gradual changes in the nuclear TF concentrations. This is effectively an amplitude modulation (AM) mechanism [bib_ref] Negative feedback that improves information transmission in yeast signalling, Yu [/bib_ref] [bib_ref] A hyper-dynamic equilibrium between promoter-bound and nucleoplasmic dimers controls NF-κB-dependent gene activity, Bosisio [/bib_ref] [bib_ref] Encoding NF-κB temporal control in response to TNF: distinct roles for the..., Werner [/bib_ref] [bib_ref] Noncooperative interactions between transcription factors and clustered DNA binding sites enable graded..., Giorgetti [/bib_ref] [bib_ref] Signal-dependent dynamics of transcription factor translocation controls gene expression, Hao [/bib_ref]. However, recent single-cell experiments also show pulsating signals and bursty entry of TFs into the Here, we aim to investigate the advantages and disadvantages of CM and BM (AM and FM) for encoding and decoding of constant concentrations and ramps. To build intuition, we start with a single receptor/ion channel (CM and BM). We consider concentration sensing by a linear pathway, allowing us to gain exact results for different temporal regimes (as suitable for fast signaling and slow gene regulation). To provide analytical results, we extend the single-receptor model for ramp sensing by In FM, the nuclear TF concentration is always the same during a burst, only the frequency of occurrence changes. As a consequence, the protein ratio stays constant. and Wingreen. First, we introduce an alternative mechanism to integral feedback, the incoherent feedforward loop (another common pathway motif for ramp sensing and precise adaptation . This allows us to generalize the model to more than one pathway. Second, by explicitly including the time-dependence of signaling noise, we are able to provide first-order analytical results for the accuracy of ramp sensing. Taken together, a general principle emerges, favoring BM for fast signaling and CM for slow gene regulation. Finally, we generalize to many receptors and ion channels, a far more realistic situation for biological systems, allowing us to make connection with AM and FM. While we found that
# Results
Cells sense external stimuli with cell-surface receptors and/or ion channels, which ultimately lead to changes in the concentration and dynamics of active transcription factors (TFs) inside the nucleus. Cells control the response at two different levels. Firstly, cell-surface receptors signal to regulate the activity of TFs in the cytoplasm. Secondly, inportin and exportin regulate the entry of active TFs into the nucleus, thereby regulating transcription . Here, we build a theoretical model that encodes information from an extra-cellular environment in an intra-cellular representation. We distinguish two ways of encoding this information: continuous modulation (CM) and bursty modulation (BM). Once the information is encoded, various proteins can act together to implement a response (decoding), involving regulatory networks. To provide a general analysis for arbitrary noise we first address concentration sensing in a simple linear pathway using the master equation. However, to derive analytical results for ramp sensing and pathways with feedback we apply the small-noise approximation. We finally extend these models to implement amplitude (AM) and frequency modulation (FM) for many receptors or ion channels. Accuracy is assessed by comparing the protein output noise for the different modulation schemes, assuming that the signal is decoded by the average concentration.
## Single-receptor/ion-channel model
## Following mora and wingreen [23] we build a single-receptor model that implements cm and bm.
We call the extra-cellular species c, which is encoded intra-cellularly by the signaling rate u. Assuming we are in the fast diffusion regime in which each ligand molecule can bind the receptor only once, the receptor can be in either of two uncorrelated states: on when bound and off when unbound. This allows the receptor activity, r(t), to be written mathematically as a binary response, which takes value 1 in the on state and 0 in the off state. The extra-cellular concentration c affects the unbound time intervals τ u , such that the binding rate is given by τ u −1 = k + c(t), where k + is the binding rate constant. In contrast, the bound time intervals, τ b , are exponentially distributed random numbers with average τ b −1 = k − , where k − is the unbinding rate constant, which is independent of the extra-cellular stimulus concentration (inset in . As for ion channels, some are ligand-gated or regulated by receptors, while others are voltage-gated and hence dependent on action potentials . In all these cases the stimulus affects the opening or closing times. In CM downstream proteins are produced with a constant rate α during each on time interval, which leads to a signaling rate u CM = αr(t), while in BM ζ = αk −1 − molecules are produced instantly at the moment of binding with rate u BM = ζ δ(t − t + i ), where t + i are the binding times . This choice for ζ allows a meaningful comparison of CM and BM as both produce, on average, the same amount of intracellular species.
## General approach to concentration sensing exhibits two regimes of accuracy
In order to provide a general result for arbitrary input fluctuations, we write down the chemical master equation. For simplicity, we only consider concentration sensing with c(t) = c 0 , but the model can also be applied to ramps. Furthermore, we assume a linear pathway in which the receptor/ion channel activity r directly regulates an output species with copy number n (with production rate u and degradation rate γ) . Since the receptor/ion channel activity is a two-state system (on/off), there are two resulting master equations for CM (one for each state) describing the probability of being in the on and . Schematic view of signaling and gene regulation. (A) Cartoon of S. cerevisiae in presence of extracellular calcium, considered a paradigm of bursty frequency modulation. Calcium enters through plasma-membrane ion channels and can be stored (released) in (from) vacuoles. Intracellular calcium activates calcineurin, which dephosphorylates Crz1p. Once dephosphorylated, Crz1 binds inporting Nmd5p and enters the nucleus. Exportin Msn5p subsequently removes Crz1 from the nucleus. Cytoplasmic calcium pulses may correspond to Crz1 bursts in the nucleus . Red arrows indicate movement while blue arrows stand for chemical signaling. (B) Single receptor/ion channel activity, r(t) (blue line), depends on the concentration of extra-cellular stimulus c. The signaling rate u differs between continuous (CM) and bursty modulation (BM). In CM, u is constant rate α during bound intervals, with p b the probability of being bound. In BM, ζ molecules are realized at the time of binding with τ bursts the duration between consecutive bursts (binding events). (C) Different regulatory networks. Linear pathway used for concentration sensing. Incoherent feedforward loop and integral feedback control allow chemical ramps to be sensed.
off states, i.e. p on (n, t) and p off (n, t): dp on (n, t) dt = γ(n + 1)p on (n + 1, t) + αp on (n − 1, t) + k + cp off (n, t) − (γn + α + k − )p on (n, t),
[formula] dp off (n, t) dt = γ(n + 1)p off (n + 1, t) + k − p on (n, t) − (γn + k + c)p off (n, t).(1a) [/formula]
Note that α ≥ k − , so molecules are generally produced in the on state. In BM, instead, the master equations which describe the probabilities p on (n, t) and p off (n, t) of having n proteins at time t, are given respectively by dp on (n, t) dt = γ(n + 1)p on (n + 1,
[formula] t) + k + cp off (n − ζ, t) − (γn + k − )p on (n, t), (2a) dp off (n, t) dt = γ(n + 1)p off (n + 1, t) + k − p on (n, t) − (γn + k + c)p off (n, t),(2b) [/formula]
with burst size ζ a positive integer. We solve Eqs. (1a) and (1b) with generating functions and simulate Eqs. (2a) and (2b) with the Gillespie algorithm (see Materials and Methods).
Simulations via the Gillespie algorithm show different outcomes for fast (small-noise approximation limit, [fig_ref] Figure 4: The two regimes in the linear pathway model based on the master... [/fig_ref] and slow [fig_ref] Figure 4: The two regimes in the linear pathway model based on the master... [/fig_ref] dynamics of the receptor. For fast switching (k + c 0 , k − ≫ γ), for both CM and BM, the probability has an unimodal distribution [fig_ref] Figure 4: The two regimes in the linear pathway model based on the master... [/fig_ref]. On the other hand, in the slow switching regime (k + c, k − ≪ γ), the probability distribution becomes bimodal for CM and unimodal with a long tail for BM, leading to drastically increased noise [fig_ref] Figure 4: The two regimes in the linear pathway model based on the master... [/fig_ref]. The unimodal distribution for BM, which is simply due to the use of infinitely short pulses, would become bimodal for finite width pulses.
In order to classify the different dynamics and to compare CM and BM for arbitrary noise, we require information on the probability distribution of n output proteins. In particular, we study the average, variance and skewness (the latter is encoded in the third moment) of the distribution for both CM and BM. Constraining the average output of CM and BM to be the same [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref] , we identify two regimes for fast dynamics: k + c 0 < k − [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref] and k + c 0 > k − [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref]. Specifically, for k + c 0 < k − , BM is more accurate [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref] , while CM is generally more accurate when k + c 0 > k − [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref] , inset), except for minimal burst size (ζ = 1). However, for slow dynamics (and hence large noise), CM is always more accurate than BM. The study of the third moment shows that, for slow switching and hence bimodality, BM has large asymmetry [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref].
These observations can be explained as follows, using the fact that the receptor/ion channel can only detect information from the extra-cellular environment during unbound (off ) time intervals, as the extra-cellular stimulus only affects the binding rate . For fast dynamics, the two regimes can be understood by comparison with maximum-likelihood estimation (MLE), the most accurate strategy for encoding . MLE estimates the ligand concentration c ML = k −1 + τ u −1 from the average unbound time interval τ u . The bound time intervals are discarded as they only contribute noise . BM, which produces fixed-size bursts at the times of binding, approaches MLE when the bound intervals are shorter than the unbound intervals. In this case, the times of the bursts effectively estimate the unbound time intervals , bottom) and BM is more accurate than CM. However, when the bound intervals are longer than the unbound intervals, BM cannot estimate the unbound time intervals anymore and becomes less accurate than CM. Since CM produces protein during the bound intervals, it signals according to the average receptor activity . Hence, CM effectively contains information on both bound and unbound intervals, and thus can still provide a reasonable estimate of unbound time intervals. An interesting exception is αk −1 − = ζ = 1, for which BM becomes slightly more accurate than CM. In the latter case, since the rate of protein production during a bound interval in CM is very low, there is uncertainty as to whether CM actually produces protein or not, which reduces its accuracy. In contrast, for slow switching the burst size needs to increase since BM produces the same level of protein as CM. Hence, BM is always less accurate than CM, independent of whether bound or unbound time intervals are longer. While we analytically demonstrate the connection with MLE for fast dynamics in the next section, an extended discussion without comparison to MLE can be found in S1 Text and S1-S3 Figs. (A-B) fast (k + c 0 = 20s −1 , k − = 100s −1 , γ = 0.1s −1 , α = 100s −1 , ζ = 1) and (C-D) slow (k + c 0 = 0.01s −1 , k − = 0.05s −1 , γ = 1s −1 , α = 25s −1 , ζ = 500) switching. (A,C) Protein number as a function of time from Gillespie simulations for CM (blue lines) and BM (red lines). (B) The probability distribution for n target proteins is unimodal for both AM (blue) and FM (red). (D) The probability distribution is bimodal for AM (blue) and remains unimodal for BM (red) but with a long tail in the slow switching regime.
[formula] p b = τ b /( τ b + τ u ) [/formula]
Small-noise approximation to ramp sensing confirms two regimes for fast dynamics To further investigate fast dynamics, we extend an analytical model for ramp sensing in the small-noise approximation . Considering the single-receptor described in ,B, we linearize the system by averaging over a time much larger than the binding and unbinding times. We further assume exponential distributions for τ b and τ u so that (
[formula] δτ b ) 2 = τ b 2 = k −2 − and (δτ u ) 2 = τ u 2 = (k + c(t)) −2 , [/formula]
where c(t) increases only very slowly with time (see below). Hence, signaling noise arises in CM due to variable bound time intervals (ignoring stochastic production of protein during bound intervals), while in BM the binding times (bursting times) vary. Without loss of generality, we set α = k − , which is equivalent to
[formula] , f = k + c 0 /(1 + k + c 0 /k − ). [/formula]
(Insets) Magnification of small-noise approximation region (fast switching). Analytical results for CM (blue) and numerical results for BM (red) as function of the frequency of binding events (logarithmic scale). Two regimes are shown: k − = 10 k + c 0 (α = 100s −1 , γ = 1s −1 , ζ from 1000 to 1) (left column) and k − = 0.1 k + c 0 (α = 10s −1 , γ = 1s −1 , ζ from 1000 to 1) (right column). Averages from CM and BM are constrained to be equal, i.e. ζ = αk −1 − . Variances of CM and BM exhibit two different regimes for fast switching: for k + c 0 < k − BM is more accurate than CM (inset in C), while for k + c 0 > k − CM is generally more accurate (inset in D), except for ζ = 1. Third moments show that, for large noise, the probability distributions become asymmetric. ζ = 1. Hence, as we show in S1 Text, for averaging time much longer than k −1 − and (k + c(t)) −1 , the average and autocorrelation (variance) of u(t) are given by [23]
[formula] u(t) = k + c(t) 1 + k + c(t)/k − ,(3)δu(t)δu(t ′ ) = g k + c(t) (1 + k + c(t)/k − ) 3 δ( t − t ′ ),(4) [/formula]
with δu(t) = 0 and
[formula] g = 1 + (δτ b ) 2 / τ b 2 = 2 CM 1 + (δτ b ) 2 / (δτ u ) 2 = 1 + [k + c(t)/k − ] 2 BM.(5) [/formula]
Note that only the variance differs between CM and BM. In particular, in Eq. (5) the ratio k + c(t)/k − determines whether g is larger in BM or CM, which ultimately determines which scheme leads to the least noise. BM has the lower noise only when k + c(t) < k − , i.e. when τ b < τ u . In particular, in the limit of fast unbinding (k + c(t) ≪ k − ), the signaling noise for CM is twice as large as for BM. Sensing temporal ramps, i.e. the change of concentration with time, is crucial for locating nutrients and avoiding toxins. We start by considering a stimulus whose concentration is constant for t < 0 and increases linearly and slowly in time after t = 0:
[formula] c(t) = c 0 t < 0 c 0 + c 1 t t ≥ 0,(6) [/formula]
for constants c 0 and c 1 with c 1 t ≪ c 0 . By applying Eq. (6) to Eqs. , the signaling rate can be rewritten to first order as
[formula] u(t) ≃ u 0 + u 1 t + δu t ≥ 0 u 0 + δu t < 0,(7) [/formula]
where u 0 , u 1 are functions of c 0 and c 1 , and δu is the noise described by δu(t)δu(t ′ ) (given in S1 Text). The condition c 1 t ≪ c 0 is necessary so that u behaves linearly in time with u 1 t ≪ u 0 . Under this condition, the factor g BM of Eq. (5) becomes
[formula] g BM ≃ 1 + k 2 + c 2 0 k 2 − g * BM + 2k 2 + c 0 c 1 k 2 − t,(8) [/formula]
where g BM is given by g * BM for a constant external concentration. We now assume that the extra-cellular stimulus is encoded in the signaling rate u which affects the production of two output proteins with concentrations x and y. Specifically, we compare the output noise of x and y between CM and BM using the incoherent feedforward , middle) and integral feedback , right) loops.
## Incoherent feedforward loop
The incoherent feedforward loop is a network motif in which u directly affects two outputs x and y, while y inhibits x . The loop provides precise adaptation to a step-change in stimulus and can also be used for ramp sensing. Mathematically, we use the following two coupled stochastic differential equations,
[formula] dx dt = k x f (u) g(y) − x ,(9)dy dt = u − k y y,(10) [/formula]
where k x is the rate constant for production and degradation of x, while k y is the rate constant for degradation of y, and f (u) and g(y) are specified functions. In order to have adaptation the variable y needs to evolve slower than x, which requires k x > k y . Here we choose f (u) = e bu and g(y) = e bkyy , where constant b has units of time. This allows us to obtain an analytic solution (see S1 Text for details).
## Integral feedback loop
The integral feedback loop [23] is another network motif for precise adaptation and ramp sensing. Here, u affects x only (the main output), while x activates y and y inhibits x . The general equations for this model are given by
[formula] dx dt = uf (y) − k x x,(11)dy dt = k y (x − 1) ,(12) [/formula]
where k x is the rate constant for degradation of x, k y is the rate constant for production and degradation of y satisfying k x > k y , and f (y) is a monotonically decreasing function of y. Specifically, we choose f (y) = e −by , where b is a dimensionless constant. This again produces an analytic solution (see S1 Text for details).
## Small-noise approximation
To analytically solve Eqs. [bib_ref] Noncooperative interactions between transcription factors and clustered DNA binding sites enable graded..., Giorgetti [/bib_ref] and (10) for the incoherent feedforward loop, and Eqs. (11) and (12) for the integral feedback loop, we linearize these equations within the small-noise approximation, and assume that we are in the fast-switching regime. This allows us to find analytic solutions in a particular time window and under certain conditions which we define in S1 Text. Specifically, for the incoherent feedforward loop in the small-ramp regime, the average values of u(t) , x(t) and y(t) are determined by the differential equations Eqs. [bib_ref] Signal-dependent dynamics of transcription factor translocation controls gene expression, Hao [/bib_ref]. Although there are no steady states for ramps, x(t) and y(t) show time-dependent stable solutions
[formula] x(t) = e bu 1 ky ,(13a)y(t) = u 0 k y − u 1 k 2 y + u 1 t k y .(13b) [/formula]
Introducing x = x + δx and y = y + δy into Eqs. [bib_ref] Noncooperative interactions between transcription factors and clustered DNA binding sites enable graded..., Giorgetti [/bib_ref] and (10) with subsequent linearization the variance of the target-protein copy numbers can be derived (see Materials and Methods). To first order in small-ramp parameters the variances of x for both types of modulation are
[formula] (δx(t)) 2 CM = ∆ g CM u 0 − 1 − 2c 0 k + /k − k x + k y u 1 + g CM (1 − 2k + c 0 /k − )u 1 t ,(14a)(δx(t)) 2 BM = ∆ g * BM u 0 − 1 − 2c 0 k + /k − + 3c 2 0 k 2 + /k 2 − 2k y u 1 + 1 − 2k + c 0 /k − + 3k 2 + c 2 0 /k 2 − u 1 t , (14b) where ∆ = b 2 k 2 x e 2bu [/formula]
1 ky 2(kx+ky)(1+k+c0/k−) 2 , and g CM and g * BM are parameters discussed in Eqs. and [bib_ref] Encoding NF-κB temporal control in response to TNF: distinct roles for the..., Werner [/bib_ref]. The corresponding results for species y are provided in Eqs. (S56) and (S58), and plots for species x and y are shown in ,D.
Consistent with the master equation, these results show again two regimes: ramp sensing is more accurate for BM if k + c 0 < k − , while CM is more accurate otherwise. For a constant environment (zeroth-order with c 1 = u 1 = 0) the regime is largely determined by the factor g. If k + c 0 < k − , . Two regimes in incoherent feedforward loop based on the small-noise approximation. Output noise, i.e. relative variance of x (top) and y (bottom), as function of the non-dimensional ramp time u 1 t/u 0 for k + c 0 < k − i.e. τ b < τ u (left) and k + c 0 > k − i.e. τ b > τ u (right). CM and BM are shown by blue and red lines respectively. (A,B) BM is more accurate than AM for k + c 0 = 10 7 s −1 and k − = 6.7 × 10 7 s −1 . (C,D) CM is more accurate then BM for k + c 0 = 10 7 s −1 and k − = 6.7 × 10 6 s −1 . Remaining parameters: k + c 1 = 10 5 s −2 , k x = 5s −1 and k y = 10s −1 .
[formula] g BM = 1 + (δτ b ) 2 / δ(τ u ) 2 < 2 (see Eq. (5)) [/formula]
, and BM is more accurate than CM with . This is because the variability of the bound intervals (δτ b ) 2 can be eliminated in BM (but not in CM), and the unbound intervals are well approximated by the duration between bursts (τ bursts in . For k + c 0 ≪ k − , BM effectively implements MLE. In contrast, CM is more accurate for k + c 0 > k − , where g CM = 2 and g BM > 2 . This is because BM contains no information on unbound time intervals, while CM still contains some information through the probability of being bound (p b in . These results also apply to ramp sensing since the accuracy of the downstream proteins (decoding) relates again to the factor g and hence to the ratio between the bound and unbound time intervals. The integral feedback loop in Eqs. (11) and (12) shows very similar behavior (provided in S1 Text). The validity of our analytical results are confirmed by simulations of the stochastic differential equation for both pathways in S4 [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref] AM is more accurate than FM for multiple receptors/ion channels To address the question of whether AM or FM is more accurate in encoding and decoding, we consider a straightforward generalization to multiple receptors (or ion channels) (see S1 Text and S6 . AM can be obtained by considering unsynchronized CM receptors. In contrast, the experimentally observed sporadic bursts of nuclear translocation [bib_ref] Signal-dependent dynamics of transcription factor translocation controls gene expression, Hao [/bib_ref] and hence FM might be explained by synchronized receptors that individually operate with BM.
[formula] g CM = 1 + (δτ b ) 2 / τ b 2 = 2 ( [/formula]
For N unsynchronized (us) receptors, the resulting average and variance of the signaling rate are in terms of the single-receptor quantities. Consequently, the relative variance, given by the variance divided by the average-squared, scales with the inverse of the number of receptors (N ). On the other hand, for N synchronized (s) receptors, the average and variance of the signaling rate are given respectively by
[formula] u(t) us N = N u(t) 1 and δu(t)δu(t ′ ) us N = N δu(t)δu(t ′ )u(t) s N = N u(t) 1 and δu(t)δu(t ′ ) s N = N 2 δu(t)δu(t ′ ) 1 . [/formula]
The relative variance is now independent of N . Hence, unsynchronized receptors (AM) have a reduction of noise by a factor N compared to synchronized receptors (FM).
For slow dynamics, or fast dynamics with k + c > k − , CM is generally more accurate than BM (at least for ζ > 1), and with N receptors, AM is more accurate than FM by an even larger margin. In contrast, for fast dynamics with k + c < k − , BM is more accurate than CM by at most a factor of 2 (Eq. (5)). But since AM is N times more accurate than CM, AM becomes more accurate for encoding than FM for more than two receptors. Since our results from the previous sections show that larger signaling noise leads to larger output noise, the same rule emerges for decoding.
From a physical point of view, how can receptors act in a synchronized fashion? Receptors may be coupled by adaptor proteins or elastic membrane deformations, allowing them to act cooperatively . In conclusion, while for fast dynamics (small-noise approximation) BM can be more accurate than CM up to a factor of two, two receptors/ion channels are sufficient for AM to become more accurate than FM. Since cells have thousands of receptors and ion channels, AM becomes the most accurate modulation scheme.
# Discussion
Cellular responses to extra-cellular stimuli involve both encoding the external stimuli by internal signals (which is normally fast) and subsequently decoding via the regulation of protein levels (which is normally much slower). The internal representation of the external signal falls into two broad categories: continuous/amplitude modulation (CM/AM), where bound receptors continually signal and the internal concentration itself encodes the external signal, and bursty/frequency modulation (BM/FM), where receptors only signal when first bound and the signal is encoded in the frequency of peaks. Here, we compared the output noise for both types of modulation in the presence of a constant and a linearly increasing (in time) external concentration. Besides considering a linear pathway, we compared two nonlinear network motifs: the incoherent feedforward loop and the integral feedback loop. These loops are ubiquitous in biological systems. For example, the incoherent feedforward loop is found in chemotactic adaptation of eukaryotes [40] and transcription networks in bacteria , and the integral feedback loop is found in chemotactic adaptation of bacteria [25, 47] and in eukaryotic olfactory and phototransduction pathways .
We found that, for a single receptor or ion channel, BM can be more accurate than CM for fast dynamics. This situation can occur when the average duration of the active on state is shorter than the average duration of the inactive off state (Figs. 5 and 6). In this case, BM effectively implements maximum-likelihood estimation, the most accurate mechanism of sensing [44]. If instead more time is spent in the on state, then CM is generally more accurate (except when the burst size is minimal, i.e one). The reason behind this effect, which we analytically prove within the small-noise approximation, is that CM has information about both the on and off states, whereas BM only knows when a switch from off to on occurs. As such, CM effectively implements Berg and Purcell's classic result of estimating ligand concentration by time averaging (see also Discussion in . In addition, we found that for slow dynamics CM is always more accurate than BM, independent of whether more time is spent in the on or off states, due to increased burst sizes [fig_ref] Figure 5: First three moments of the protein distribution in concentration sensing from the... [/fig_ref]. Taken together our results suggest that BM should be more common in signaling pathways than in gene regulation.
The generalization to multiple receptors/ion channels allows AM and FM to be compared. AM, which arises from unsynchronized CM receptors, has a reduced relative noise due to spatial averaging, while the relative noise in FM from synchronized BM receptors remains identical to the single-receptor result. (Note the observed nuclear bursts of approximately constant amplitude and duration support our FM mechanism [bib_ref] Signal-dependent dynamics of transcription factor translocation controls gene expression, Hao [/bib_ref] As a result, AM is always more accurate than FM for more than two receptors (S6 . Since cells have tens of thousands of receptors and ion channels, this implies that the reason that FM is sometimes observed in real systems must have a different origin. At least three possibilities present themselves. Firstly, FM can help to coordinate gene expression , which is particularly useful when hundreds of genes are controlled by a single transcription factor, such as during stress response [bib_ref] Genome-wide analysis of gene expression regulated by the calcineurin/Crz1p signaling pathway in..., Yoshimoto [/bib_ref] [bib_ref] Circuitry of nuclear factor κB signaling, Hoffmann [/bib_ref] [bib_ref] Msn2p, a zinc finger DNA-binding protein, is the transcriptional activator of the..., Schmitt [/bib_ref]. Secondly, FM can enhance co-localization of proteins inside the nucleus, providing another way to improve coordination of gene expression [bib_ref] A dynamical model reveals gene co-localizations in nucleus, Kang [/bib_ref]. Thirdly, as with oscillatory signals, bursts can be used to activate transcription by threshold crossing [32] while avoiding desensitization . This may then push the cell to differentiate into a new state (such as under starvation to initiate competence) [bib_ref] Bistability, bifurcations, and Waddington's epigenetic landscape, Ferrell [/bib_ref] [bib_ref] p53 dynamics control cell fate, Purvis [/bib_ref]. It is also worth noting that by using seemingly redundant isoforms (such as NFAT1 and NFAT4 during an immune response), AM and FM can be combined to enhance temporal information processing .
While providing intuitive insights, it is clear that our models are highly oversimplified versions of signaling and gene regulation in actual cells. One of the main reasons for this is that we used idealized delta-functions as pulses in BM (and hence in FM). However, for example, in the calcium stress-response pathway in Saccharomyces cerevisiae nuclear bursts of Crz1p are on average two minutes long [fig_ref] Figure 1: Experimental evidence for amplitude and frequency modulation [/fig_ref]. Most likely cytoplasmic calcium spikes determine the nuclear bursts (Elowitz, personal communication), but since the mechanism of calcium spiking remains poorly understood, such bursts are difficult to model. A further limitation of our models is that bursts only relate to translocation, whereas additional bursts may occur further downstream during transcription [bib_ref] Live imaging of nascent RNA dynamics reveals distinct types of transcriptional pulse..., Muramoto [/bib_ref] (e.g. due to promoter switching [24]) and translation [bib_ref] Intrinsic noise in gene regulatory networks, Thattai [/bib_ref]. Future models may need to include these details.
Our models suggest further experimental investigation in multiple areas. Firstly, the distribution of burst duration affects factor g (Eq. (5)), so that g = 2 in equilibrium for a single-step process and potentially g < 2 for an irreversible binding cycle dominated by energy dissipation [bib_ref] Thermodynamics of statistical inference by cells, Lang [/bib_ref]. These irreversible cycles are present in some ligand-gated ion channels, such as the cystic fibrosis transmembrane conductance regulator (CFTR) channels and N-Methyl-D-aspartate (NMDA) receptors. These exhibit peaked opening distributions, which can be interpreted as evidence of broken reversibility and energy consumption [bib_ref] Strict coupling between CFTR's catalytic cycle and gating of its Cl-ion pore..., Csanády [/bib_ref] [bib_ref] Coupling of permeation and gating in an NMDA-channel pore mutant, Schneggenburger [/bib_ref]. Such cases and their possible connection with accuracy need further investigation. In fact, most cellular processes rely heavily on energy consumption, including nuclear shuttling and chromosome remodeling, limiting the applicability of our equilibrium CM-receptor model. Secondly, coordination of gene expression during stress or cell-fate decisions might be another reason for implementing FM rather than AM. More quantitative experiments are needed to better understand this mechanism. Thirdly, closer inspection of Ca 2+ -independent transcription factors (as well as Ca 2+ -dependent co-regulated genes) are warranted in order to verify coordination of multiple genes . Finally, to see if bursts help jump start new cellular programs (i.e. transition into a new "attractor"), global changes in gene regulation can be monitored.
A general understanding of FM may help prevent developmental defects and human diseases. Indeed, several biomedically relevant transcription factors, such as NF-κB, p53, NFAT and ERK, show oscillatory pulsing or random bursting [bib_ref] p53 dynamics control cell fate, Purvis [/bib_ref]. In fact, the destabilization of regulatory circuits can underlie human diseases: studies suggest that the coordination of gene expression could be critical in maintaining the proper functioning of key nodes in such circuits. For example, the NFATc circuit is cooperatively destabilized by a 1.5-fold increase in the DSCR1 and DYRK1A genes, which reduce NFATc activity leading to characteristics of Down's syndrome [bib_ref] NFAT dysregulation by increased dosage of DSCR1 and DYRK1A on chromosome 21, Arron [/bib_ref]. However, ERK pulses are regulated by both AM and FM with the same dose dependence, and it remains unclear how they affect cell proliferation and the relevance to cancer .
Broadly speaking, temporal ordering (regularity or periodicity) serves at least two roles in living systems [bib_ref] More is different, Anderson [/bib_ref] : extraction of energy from the environment and handling of information. While the first role is well studied in terms of molecular motors at the single-molecule level, the second role is intellectually more difficult to understand as it requires a broader, more global understanding of cells. We believe that future work that combines single-cell experiments with ideas of collective behavior and engineering principles is most likely to be successful.
# Materials and methods
## Master-equation model for concentration sensing
The master equations for continuous modulation (CM), Eqs. (1a) and (1b), can be solved at steady state using generating functions. In particular, we derive the first three moments of the probability distribution using the general model in [bib_ref] Energetic costs of cellular computation, Mehta [/bib_ref]. When the system is in the on/off state, the production rate of species x is α on/off . The degradation rate γ is independent of the state of the system. The probability distribution of n target proteins at time t is then described by dp s (n, t) dt = γ(n + 1)p s (n + 1, t) + α s p s (n − 1, t) + ksps(n, t) − (γn + α s + k s )p s (n, t),
wheres = off (on) when s = on (off). By defining the generating functions
[formula] G s (z) = ∞ n=0 p s (n)z n ,(16) [/formula]
and using Eq. (15), a solution for G s (z) can be found, which then readily gives the moments of p(n, t).
In particular, the variance and skewness are given by
[formula] δn 2 = s (∂ z z∂ z G s (z)) z=1 − n 2 ,(17)n 3 = s [∂ z z∂ z z∂ z G s (z)] z=1 .(18) [/formula]
Full details are given in S1 Text. In order to solve the master equation for bursty modulation (BM), Eqs. (2a) and (2b), we use the Gillespie algorithm [bib_ref] A general method for numerically simulating the stochastic time evolution of coupled..., Gillespie [/bib_ref]. If the system is in the on state with n proteins at time t, it can either switch to the off state with transition rate given by k − /(k − + γn) or else remain in the on state and lose a protein by degradation. If instead the system is in the off state with n proteins at time t, it can either switch to the on state with switching rate k + c 0 /(k + c 0 + γn) and, via a burst, increase its number of proteins to n + ζ, or again remain in the same state and loose a protein by degradation. The time step between reactions, δt, is chosen from an exponential probability distribution λe −λδt , with λ equal to the total rate that at least one reaction occurs.
## Ode models for ramp sensing
The following method applies to both the incoherent feedforward and the integral feedback loop. To solve the ordinary differential equations (9-12) we linearize around stable solutions, x(t) = x(t) + δx and y(t) = y(t) + δy, and assume that small δu leads to small δx and δy. Note that when sensing a gradually changing ramp, x(t) and y(t) are not steady states. Defining X = [x(t) y(t)] T we can rewrite these equations as
[formula] dX(t) dt + M X(t) = w δu z δu ,(19) [/formula]
where the matrix M and the constants w and z are defined in S1 Text. Analytic solutions are only available when M is time-independent. As shown in S1 Text, Eq. (19) can be solved and written as an integral, which can then be evaluated with, for example, Wolfram Mathematica 8. Supporting Information Legends S1 Text. Details of analytical calculations. . Variances of CM, BM and IM exhibit two different regimes for fast switching: for k + c 0 < k − BM is the most accurate mechanism and CM the worst (inset in C) while for k + c 0 > k − CM is generally the most accurate (except for ζ = 1) and IM the worst (inset in D). Third moments show that, for large noise, the probability distributions become asymmetric. In this section we calculate the analytic solution for the master equation (Eq. ??,b in the main text) for continuous modulation (CM). For clarity, we repeat here the master equations for the on and off states: dp on (n, t) dt = γ(n + 1)p on (n + 1, t) + αp on (n − 1, t) + k + c p off (n, t) − (γn + α + k − )p on (n, t), dp off (n, t) dt = γ(n + 1)p off (n + 1, t) + k − p on (n, t) − (γn + k + c)p off (n, t),
[formula] k + c 0 < k − with k + c = 0.1 k − (α = 100s −1 , γ = 1s −1 ). (Bottom) Regime k + c 0 > k − with k + c = 10 k − (α = 10s −1 , γ = 1s [/formula]
where p on/off (n, t) is the probability that the receptor/ion channel is in the on/off state with n output proteins at time t, α and γ are the production and degradation rates respectively, and k + c and k − are the binding and unbinding rates respectively. Note that the concentration of the input species (c) is now constant. Eqs. (??,b) can be rewritten as Eq. (??) in the main text dp s (n, t) dt = γ(n + 1)p s (n + 1, t) + α s p s (n − 1, t) + ksps(n, t) − (γn + α s + k s )p s (n, t),
wheres is the on/off state when s is the off /on state, and α on = α, α off = 0, k on = k − and k off = k + c. To find the solution for the first two moments of the distribution p(n, t), we now follow Mehta and Schwab . At steady state Eq. (??) becomes Ksps(n) = −(n + 1)p s (n + 1) − A s p s (n − 1) + (n + A s + K s )p s (n),
where K s = k s /γ and A s = α s /γ. Using the generating function in Eq. (??)
[formula] G s (z) = ∞ n=0 p s (n)z n , Eq. (S2) becomes [(z − 1)∂ z − A s (z − 1) + K s ] G s (z) = Ks,(S3) [/formula]
which implies
[formula] (∂ z − A on ) G on = K off G off − K on G on z − 1 ,(S4)(∂ z − A off ) G off = K on G on − K off G off z − 1 ,(S5) [/formula]
which, when combined, gives To proceed further, it is useful to define the quantity H s (z) related to the generating function G s (z) by
[formula] (∂ z − A on ) G on = − (∂ z − A off ) G off .(G s (z) = e Asz H s (z). (S7) Using ∂ z − A s G s (z) = e Asz ∂ z H s (z),(S8) [/formula]
Eq. (S6) becomes
[formula] e Aonz ∂ z H on (z) = −e A off z ∂ z H off (z),(S9) [/formula]
which links the expressions for H on and H off . At this point the initial equation for the steady state (Eq. (S3)) becomes
[formula] (z − 1)e Asz ∂ z H s (z) + K s e Asz ∂ z H s (z) = Kse Asz ∂ z Hs(z).(S10) [/formula]
Multiplying by e As , taking the derivative with respect to z, substituting Eq. (S9), and defining ∆A s = As − A s , gives
[formula] ∂ z H s (z) − ∆A s (z − 1)∂ z H s (z) + (z − 1)∂ 2 z H s (z) − K s ∆A s H s (z) + K s ∂ z H s (z) = −KsH s (z) and hence (z − 1)∂ 2 z H s (z) + 1 − ∆A s (z − 1) + K s + Ks ∂ z H s (z) − K s ∆A s H s (z) = 0. [/formula]
Finally, changing variables to u = ∆A s (z − 1) provides
[formula] u∂ 2 z H s (u) + (1 + K s + Ks − u) ∂ z H s (u) − K s H s (u) = 0. (S11) [/formula]
This is the confluent hypergeometric equation, for which the solution in terms of confluent hypergeometric functions of the first kind is given by
[formula] H s (u) = c s 1 F 1 (K s , 1 + K s + Ks; u) ,(S12) [/formula]
with c s a constant of integration. Thus, through Eq. (S7), G s (z) = c s e Asz 1 F 1 (K s , 1 + K s + Ks; ∆A s (z − 1)) .
(S13)
To determine the constants, notice that 1 F 1 (a, b, 0) = 1 leading to G s (1) = c s e As = p s = Ks
[formula] Ks + K s , [/formula]
where p s is the average probability of being in state s. Rearranging terms, we obtain
[formula] c s = Ks Ks + K s e −As .(S14) [/formula]
Finally, the probability distribution at steady state is given by G s (z) = Kse As(z−1)
[formula] Ks + K s 1 F 1 (K s , 1 + K s + Ks; ∆K s 2 (z − 1)) .(S15) [/formula]
Having an analytic expression for the steady-state probability distribution (Eq. S15), we can now calculate the first, second and third moments, which are related to the mean, variance and skewness, respectively. The mean production of the output protein is given by the mean production in the on state multiplied by the probability to be in the on state, averaged over the whole time period. For such a two-state system p on = K off K off +Kon and p off = 1 − p on . Therefore, the mean number of proteins is given by
[formula] n = (A on − A off ) p on + A off = α γ k + c/k − 1 + k + c/k − .(S16) [/formula]
To calculate the variance, we use the following property of the generating function:
[formula] (δn) 2 = s (∂ z z∂ z G s (z)) z=1 − n 2 .(S17) [/formula]
Proof.
(δn) 2 = s n
[formula] n 2 p s (n) − n = s n n 2 p s (n) z=1 − n = s n z∂ z (np s (n)z n ) n 2 p s (n) z=1 − n = s [z∂ z z∂ z G s (z)] z=1 − n = s [∂ z z∂ z G s (z)] z=1 − n . [/formula]
Using common properties of hypergeometric functions, the analytical solution for the variance is (δn)
[formula] 2 = n + p on p off (∆A s ) 2 1 + K s + Ks = n + α 2 γ (γ + k − + k + c) k + c/k − (1 + k + c/k − ) 2 .(S18) [/formula]
For details, see the full calculation in the SI of .
## Third moment
In order to understand more about the symmetry of the probability distribution, we calculate the third moment at steady state. As in Eq. (S17) the third moment can be found via generating functions as
[formula] n 3 = s [∂ z z∂ z z∂ z G s (z)] z=1 = s 3z∂ 2 z G s (z) + z 2 ∂ 3 z G s (z) + ∂ z G s (z) z=1 , (S19) where [1] s ∂ z G s (z) z=1 = n , (S20) s z∂ 2 z G s (z) z=1 = p on (A on ) 2 + p off (A off ) 2 − p on p off ∆A s (K on + K off ) 1 + K s + Ks = α 2 (k + c/k − ) γ 2 (1 + k + c/k − ) + α(k + c) γ(1 + k + c/k − )(γ + k − + k + c) . (S21) [/formula]
Thus, only s z 2 ∂ 3 z G s (z) z=1 needs to be calculated. The result is
[formula] s z 2 ∂ 3 z G s (z) z=1 = (A on ) 3 p on + (A off ) 3 p off − p on p off (∆A on ) 2 (K on + K off ) 1 + K on + K off × (3 + K on + 2K off ) A on 2 + K on + K off + (3 + 2K on + K off ) A off 2 + K on + K off = α 3 (k + c/k − ) γ 3 (1 + k + c/k − ) − α 3 (k + c/k − ) (3γ + k − + 2k + c) γ 3 (1 + k + c/k − ) (γ + k − + k + c) (2γ + k − + k + c) .(S22) [/formula]
By combining Eqs. (S57)-(S59) as indicated in Eq. (S56), we obtain the analytic expression for the skewness of our system.
Proof. From the definition of confluent hypergeometric functions of the first kind 1 F 1 (a, b; z) = ∞ n=0 a n b n n! z n , with a n = a(a + 1)(a + 2)...(a + n − 1), b n = b(b + 1)(b + 2)...(b + n − 1), and the fact that
[formula] z∂ z 1 F 1 (a, b; z) = z a b 1 F 1 (a + 1, b + 1; z),(S23) [/formula]
we obtain
[formula] ∂ z G s (z) = Ks Ks + K s A s e As(z−1) 1 F 1 (·, ·, ∆A s (z − 1)) + ∆A s K s 1 + K s + Ks e As(z−1) 1 F 1 (+, +, ∆A s (z − 1)) ,(S24) [/formula]
where ∆A s = As − A s and 1 F 1 (+, +, z) = 1 F 1 (a + 1, b + 1, z). We now need to calculate s z 2 ∂ 3 z G s (z) z=1 . Using Eq. (S21) we find that
[formula] z 2 ∂ 3 z G s (z) z=1 = z 2 Ks Ks + K s ∂ z (A s ) 2 1 F 1 (·, ·, ∆A s (z − 1)) + 2∆A s K s A s 1 + K s + Ks 1 F 1 (+, +, ∆A s (z − 1)) + (∆A s ) 2 K s (K s + 1) (1 + K s + Ks)(2 + K s + Ks) · 1 F 1 (++, ++, ∆A s (z − 1)) e As(z−1) z=1 . (S25) [/formula]
Differentiating and using property (S23), we obtain
[formula] z 2 ∂ 3 z G s (z) z=1 = z 2 Ks Ks + K s e As(z−1) (A s ) 3 1 F 1 (·, ·, ∆A s (z − 1)) + 3∆A s K s (A s ) 2 1 + K s + Ks 1 F 1 (+, +, ∆A s (z − 1)) + 3(∆A s ) 2 K s (K s + 1) K s (1 + K s + Ks)(2 + K s + Ks) 1 F 1 (++, ++, ∆A s (z − 1)) + (∆A s ) 3 K s (K s + 1) (K s + 2) (1 + K s + Ks)(2 + K s + Ks)(3 + K s + Ks) 1 F 1 (+ + +, + + +, ∆A s (z − 1)) z=1 . (S26) [/formula]
Evaluating at z = 1, we obtain
[formula] z 2 ∂ 3 z G s (z) z=1 = Ks Ks + K s (A s ) 3 + ∆A s K s 1 + K s + Ks 3 (A s ) 2 + ∆A s (K s + 1) 2 + K s + Ks 3A s + ∆A s (K s + 2) 3 + K s + Ks . (S27) [/formula]
Finally, summing on the possible states of s we arrive at Eq. (S22).
## Small-noise approximation to ramp sensing
## Input noise
In the Model section of the main text, we built a model for a single receptor/ion channel that encodes information from an cell-external environment in some cell-internal degrees of freedom. Similarly to , we assume that the receptor/ion channel activity (r(t)) is a two state system: on with r = 1 when the receptor is bound or the channel open, and off with r = 0 when the receptor is not bound or the channel is closed. The external concentration (c(t)) is assumed to affect the unbound/closed time interval τ u = [k + c(t)] −1 but not the bound/open time internal τ b = k −1 − , where k + and k − are both constants. Both interval durations are assumed to be independent, exponentially distributed random variables. The independence of binding and unbinding (or equivalently of opening and closing) means that the probability of a molecule binding the receptor a second time is negligible. We therefore assume the system to be in the fast diffusion regime.
The signaling rate, called u, implements two different mechanisms of encoding, either continuous (CM) or bursty (BM) modulation. CM and BM ultimately correspond to amplitude (AM) and frequency (FM) modulation, respectively, when generalized to multiple receptors/ion channels as explained in the Results section of the main text. In CM the proteins are produced with a constant rate α during the binding time. On the other hand, for BM a burst of ζ proteins is realized at the time of binding, so
[formula] u(t) = α r(t) for CM ζ δ(t − t + i ) for BM,(S28) [/formula]
where αk −1 − = ζ and t + i the binding times. By taking the average of the rate u(t) over a timet much longer than both the average bound time, τ b = k −1 − , and the average unbound time, τ u = [k + c(t)] −1 , but shorter than the time during which the external concentration changes, we obtain for further details). Importantly,
[formula] u(t) = ζ k + c(t) 1 + k + c(t)/k − , (S29) δu(t) = 0, δu(t)δu(t ′ ) = gζ 2 k + c(t) (1 + k + c(t)/k − ) 3 δ( t − t ′ ), (S30) g = 2 for CM 1 + [k + c(t)/k − ] 2 for BM,(S31)g BM < g CM since k + c(t) < k − , i.e. τ b < τ u . [/formula]
By considering an external concentration given by Eq. (??) in the main text,
[formula] c(t) = c 0 + c 1 t, t ≥ 0 c 0 , t < 0, with c 1 t ≪ c 0 , Eq. (S29) becomes u(t) = ζ u 0 + u 1 t + δu(t), t ≥ 0 u 0 + δu(t), t < 0,(S32) [/formula]
where we assume δu(t) ≪ u 0 + u 1 t and
[formula] u 0 = k + c 0 (1 + k + c 0 /k − ) ,(S33)u 1 = k + c 1 (1 + k + c 0 /k − ) 2 , (S34) δu(t)δu(t ′ ) t,t ′ ≥0 = δ(t−t ′ ) (1+k+c0/k−) 2 [g CM u 0 + g CM (1 − 2k + c 0 /k − ) u 1 t] for CM δ(t−t ′ ) (1+k+c0/k−) 2 g * BM u 0 + 1 − 2k + c 0 /k − + 3k 2 + c 2 0 /k 2 − u 1 t for BM, (S35) δu(t)δu(t ′ ) t,t ′ <0 = gu 0 (1 + k + c 0 /k − ) 2 δ(t − t ′ ).(S36) [/formula]
Here g CM = 2 and g * BM = 1 + (k + c 0 /k − ) 2 (cf. Eq. (??) in the main text). Again for ζ = 1, this becomes Eq. (??) in the main text. From Eqs. (S32)-(S36), the constant (t < 0) and ramp (t ≥ 0) regimes for the external species c are encoded in the rate u in the corresponding regimes since the condition c 1 t ≪ c 0 ensures u 1 t ≪ u 0 . However, to satisfy condition δu ≪ u for both CM and BM, a new condition is needed:
[formula] g CM/BM k + c 0 (1 + k + c 0 /k − ) ≪ τ c ,(S37) [/formula]
which implies
[formula] k − , k + c 0 ≫ τ −1 c .(S38) [/formula]
Here, we have introduced the correlation time of white noise, τ c , corresponding to the δ-function used in Eqs.(S35) and (S36). Note that condition in (S38) restricts our study to the fast switching regime. Finally, the signaling rate u in the constant regime has one small term δu/u 0 of order δ which is defined by Eqs. [fig_ref] Figure 2: Advantages and disadvantages of amplitude and frequency modulation [/fig_ref]. Instead, small-ramp regime u contains two small terms: the small-ramp term u 1 t/u 0 of order ǫ and the small noise term δu/u 0 which now has a correction to order δ coming from the small ramp (order ǫ). With these definitions, from Eqs. (S32)-(S35) the rate u in the small ramp regime has two small corrections to the constant rate u 0
[formula] u(t) = ζu 0 1 + u 1 t u 0 o(ǫ) + δu u 0 o(δ(1+ √ ǫ)) ,(S39) [/formula]
In order to linearize around linear solutions, we further assume that the small-noise amplitude is smaller than the small ramp. As a result, o(δ) ∼ o(ǫ x ) with x > 1, which means that
[formula] g(1 + k + c 0 /k − )k + c 0 < τ c (k + c 1 t) 2 . (S40) [/formula]
Note that for simplicity, both in the following sections and in the main text, we set ζ = 1.
## Output noise in incoherent feedforward loop
[formula] dx dt = k x e b(u−kyy) − x ,(S41)dy dt = u − k y y.(S42) [/formula]
Here, b is a constant introduced to maintain the exponent unitless, and k x and k y are rate constants for x and y. This system of equations performs exact adaptation. The steady-state solution in the constant regime (t < 0 in Eq. (S32)) is
[formula] x(t) = 1, y(t) = u 0 k y ,(S43) [/formula]
which sets the initial conditions x(0) = 1 and y(0) = u 0 /k y for Eqs. (S41) and (S42) in the ramp regime (t ≥ 0 in Eq. ??). With these initial conditions, the solutions for t ≥ 0 can be written as
[formula] x(t) = e −kxt k x e bu 1 ky t 0 dt ′ e kxt ′ − bu 1 ky e −ky t ′ ,(S44)y(t) = u 0 k y − u 1 k 2 y + u 1 k y t + u 1 k 2 y e −kyt ,(S45) [/formula]
where the integral in the expression for x(t) cannot be solved analytically. However, by assuming that the integral starts from time ǫ ≫ k −1 y , Eq. (S44) becomes
[formula] x(t) ≃ e bu 1 ky + e −kx(t−ǫ) x(ǫ) − e bu 1 ky . [/formula]
Finally, by considering t such that t − ǫ ≫ k −1 x and without exceeding the small-ramp regime (e.g. k x,y ≫ 1 and ǫ small), the solution becomes Eqs. in the main text,
[formula] x(t) = e bu 1 [/formula]
ky ,
[formula] y(t) = u 0 k y − u 1 k 2 y + u 1 k y t. [/formula]
These solutions match numerical results shown in S4A Fig. Note that the time interval over which these solutions are valid extends from a time larger than the transient time to around a time that does not exceed the small-ramp regime. These criteria also set the regime of validity for our next results.
## Output variances in ramp sensing
Above we gave the average solutions for the incoherent feedforward loop, both in the constant regime (t < 0, Eq. (S43)) and in the ramp regime (t ≥ 0, Eqs. [fig_ref] Figure 4: The two regimes in the linear pathway model based on the master... [/fig_ref]. Now we want to linearize the equations around these solutions in order to obtain information about the noise. We assume that the input noise (δu) is smaller than the ramp, Eqs. (S38) and (S40), which translates into small output noise (δx and δy). In addition to these assumptions, we also assume b ∼ u −1 0 in order to ensure b(δu − k y δy) ≪ 1. Hence, in the ramp regime, the differential equations for δx and δy become
[formula] d(δx) dt = k x be u 1 b ky (δu − k y δy) − δx , (S47) d(δy) dt = δu − k y δy.(S48) [/formula]
By defining X(t) = δx δy , Eqs. (S47) and (S48) can be rewritten in a compact way for both the constant and small-ramp regimes as
[formula] dX(t) dt + M (t)X(t) = A(t)δu(t) δu(t) ,(S49) [/formula]
where
[formula] M (t) = k x k x k y b 0 k y , t < 0 k x k x k y b e bu 1 ky 0 k y , t ≥ 0,(S50) [/formula]
and
[formula] A(t) = k x b, t < 0 k x b e bu 1 ky , t ≥ 0. (S51) [/formula]
Note that M (t) (for t > 0) is independent of time, which allows Eq. (S49) to be solved analytically. This is due to our choice of f (u) and g(u) in Eqs. (??) and (??) in the main text. For the constant input regime, t < 0, the solutions for CM and BM are
[formula] (δx(t)) 2 = g CM/BM b 2 k 2 x u 0 2(k x + k y )(1 + k + c 0 /k − ) 2 ,(S52) [/formula]
(δy(t)
[formula] ) 2 = g CM/BM u 0 2k y (1 + k + c 0 /k − ) 2 ,(S53) [/formula]
where g CM = 2 and g BM = g * BM = 1 + (k + c 0 /k − ) 2 . Hence, in the constant regime the output noise for BM is lower than the output noise for CM since k + c 0 < k − .
For the small-ramp regime, t ≥ 0, Eq. (S49) is analytically solvable for t ≫ k −1 y by evaluating the integral from time ǫ ≫ k −1 y to some time t that does not exceed the small-ramp approximation (as discussed for the average solutions). With these assumptions and by using an appropriate integrating factor, the solution for X(t) is
[formula] t ǫ e Mt ′ dX(t ′ ) dt ′ dt ′ + t ǫ e Mt ′ M X(t ′ )dt ′ = t ǫ e Mt ′ Aδu(t ′ ) δu(t ′ ) dt ′ , X(t) = e −Mt t ǫ e Mt ′ Aδu(t ′ ) δu(t ′ ) dt ′ + e −M(t−ǫ) X(ǫ). [/formula]
However, for t − ǫ ≫ k −1 x,y (within the limit for t and ǫ as discussed for Eqs. the solution is
[formula] X(t) = e −Mt t ǫ e Mt ′ Aδu(t ′ ) δu(t ′ ) dt ′ .(S54) [/formula]
By using matrix diagonalization, expressing the noise in u by a delta function in time (Eq. S35), and integrating Eq. (S54) for X(t) 2 , we find analytical solutions for the variances. The results for CM are
[formula] (δx(t)) 2 CM = b 2 k 2 x e 2bu 1 ky 2(k x + k y )(1 + k + c 0 /k − ) 2 g CM u 0 − g CM (1 − 2k + c 0 /k − ) k x + k y u 1 + g CM (1 − 2k + c 0 /k − )u 1 t , (S55) (δy(t)) 2 CM = 1 2k y (1 + k + c 0 /k − ) 2 g CM u 0 − g CM (1 − 2k + c 0 /k − ) ky u 1 + g CM (1 − 2k + c 0 /k − )u 1 t ,(S56) [/formula]
where g CM = 2. For BM, the g BM parameter (see Eq. (??)) affects the integration, and the results are
[formula] (δx(t)) 2 BM = b 2 k 2 x e 2bu 1 ky 2(k x + k y )(1 + k + c 0 /k − ) 2 g * BM u 0 − 1 − 2k + c 0 /k − + 3k 2 + c 2 0 /k 2 − k x + k y u 1 + 1 − 2k + c 0 /k − + 3k 2 + c 2 0 /k 2 − u 1 t ,(S57) [/formula]
(δy(t))
[formula] 2 BM = 1 2k y (1 + k + c 0 /k − ) 2 g * BM u 0 − 1 − 2k + c 0 /k − + 3k 2 + c 2 0 /k 2 − 2ky u 1 + 1 − 2k + c 0 /k − + 3k 2 + c 2 0 /k 2 − u 1 t ,(S58) [/formula]
where g * BM = 1 + (k + c 0 /k − ) 2 . Note that for c 1 = u 1 = 0, the solutions coincide with the solutions for the constant regime. Furthermore, by comparing the time-dependent terms, CM is noisier than BM when k + c 0 < 1 3 k − . In our model this is due to the input noise (cf. Eq. S35). These two regimes in which BM is less noisy and hence more accurate than CM depend on the ratio of binding and unbinding rates as shown in S4B,C Clearly the analytic solutions match numerical simulations with noise. All these calculations were done using Wolfram Mathematica 8, while all the simulations of the stochastic differential equations (Eqs. S41 and S42) were done using the Euler method in MATLAB.
## Output noise for integral feedback loop
A similar approach can be applied to the integral feedback loop given by Eqs. (??) and (??) in the main text, shown here for clarity with f (y) = e −by :
[formula] dx dt = ue −by − k x x,(S59)dy dt = k y (x − 1) .(S60) [/formula]
Assuming that u is given by Eq. (S32), this system does not have analytic solutions in the ramp regime. However, in the small-ramp regime it is possible to linearize around the solutions of the constant regime. Hence, x(t) = x 0 +ǫ x (t) and y(t) = y 0 + ǫ y (t), where x 0 = 1 and y 0 = ln (u0/kx) b are the solutions for the constant regime with u = u 0 . Note that the condition k x < u 0 is required. By linearization, Eqs. (??) and (??) become
[formula] dǫ x (t) dt = k x u 1 t u 0 − k x ǫ x (t) − k x bǫ y (t) − bk x u 1 u 0 ǫ y (t)t,(S61)dǫ y (t) dt = k y ǫ x (t).(S62) [/formula]
Combining both equations and neglecting the second-order term u 1 t/u 0 ǫ y (t), it possible to find a second-order differential equation for ǫ x (t), given by
[formula] d 2 ǫ x (t) dt 2 + k x dǫ x (t) dt + bk x k y ǫ x (t) = k x u 1 u 0 . (S63) [/formula]
The solution is
[formula] ǫ x (t) = u 1 bk y u 0 + C 1 exp − k x 2 − k 2 x 4 − bk x k y t + C 2 exp − k x 2 + k 2 x 4 − bk x k y t ,(S64) [/formula]
with ǫ x (t) → u1 bkyu0 after a transient time defined by the exponential terms for any k 2 x /4 − bk x k y . Furthermore, there are two integration constants C 1 and C 2 . From Eq. (S62) we obtain ǫ y = u1 bu0 t − u1 b 2 kyu0 . Finally, the solutions of linearized Eqs. (??) and (??) in the small-ramp regime after the transient time are x(t) = 1 + u 1 bk y u 0 , (S65)
[formula] y(t) = y 0 − u 1 b 2 k y u 0 + u 1 bu 0 t.(S66) [/formula]
Within the small-noise approximation (Eq. S38), we want to find expressions for the variances. In the constant regime, u(t) = u 0 + δu(t) (t < 0 in Eq. 5) implies x(t) = 1 + δx(t) and y(t) = y 0 + δy(t). Therefore, the equations for the noise terms become
[formula] d(δx) dt = k x δu u 0 − bu 0 δy − δx , (S67) d(δy) dt = k x δx.(S68) [/formula]
Proceeding similarly to the incoherent feedforward loop, in the constant regime the solution for the variances are (δx
[formula] ) 2 = g CM/BM k x 2 (1 + k + c 0 /k − ) 2 u 0 , (S69) (δy) 2 = g CM/BM k y 2b (1 + k + c 0 /k − ) 2 u 0 ,(S70) [/formula]
with g CM = 2 and g BM = g * BM = 1 + (k + c 0 /k − ) 2 . Hence, in the constant regime, the output noise for BM is lower than the output noise for CM (since k + c 0 < k − ).
To study the system in the small-ramp regime (t > 0 in Eq. 5), we assume that the input noise (δu) is smaller than the ramp (Eqs. S38 and S40), which translates into small output noise (δx and δy), and linearize around solutions (S65) and (S66). As a result, Eq. (??) becomes
[formula] d (δx) dt = k x 1 + u 1 t u 0 + δu u 0 1 − b (ǫ y + δy) + b 2 2 (ǫ y + δy) 2 − 1 − ǫ x − δx , [/formula]
which, by using Eq. (S61), becomes
[formula] d (δx) dt = k x b δu u 0 − δy − u 1 u 0 ǫ y t − u 1 t u 0 δy − ǫ y u 0 δu + bǫ 2 y 2 + bǫ y δy − δx b + o(ǫ 3 , δ 2 , ǫ 2 δ) ,(S71) [/formula]
where we neglect third-order terms in the small ramp (o(ǫ 3 )), second-order terms in the small noise (δ 2 ) and mixedorder terms (o(ǫ 2 δ)) due to the assumption that the noise is smaller than the ramp (cf. discussion that leads to Eq. (S40)).
Defining X(t) = δx δy , Eqs. (S71) and (S60) become
[formula] dX(t) dt + M (t)X(t) = A(t)δu(t) + B(t) 0 ,(S72) [/formula]
where from Eqs. (S71) and (S60), using definitions of ǫ x and ǫ y ,
[formula] M (t) = k x k x b 1 + bu1 kyu0 −k y 0 , A(t) = kx u0 1 + u1 kybu0 − u1 [/formula]
where the term e −M(t−ǫ) X(ǫ) is negligible for (t − ǫ) ≫ k −1 x,y . To calculate the variances we square Eq. (S73). Using Eqs. (S5)-(S9), the results for (δx(t)) 2 to first-order in the small-ramp parameters are
[formula] (δx(t)) 2 CM = kx 2 (1 + k + c 0 /k − ) 2 u 0 g CM − g CM (1 + 2k + c 0 /k − ) u 1 t u 0 + g CM 2k x + bk y (1 + 2k + c 0 /k − ) u 1 bk y k x u 0 , (S74) (δx(t)) 2 BM = kx 2 (1 + k + c 0 /k − ) 2 u 0 g * BM − 1 + 2k + c 0 /k − − k 2 + c 2 0 /k −2 − u 1 t u 0 + bk y 1 + 2k + c 0 /k − − k 2 + c 2 0 /k −2 − + 2k x g * BM u 1 bk x k y u 0 . (S75) [/formula]
Similarly the results for (δy) 2 are (δy(t)) 2
[formula] CM = k y 2b (1 + τ b k + c 0 ) 2 u 0 g CM − g CM (1 + 2k + c 0 /k − ) u 1 t u 0 + g CM (kx(3 + 2k + c 0 /k − ) + 2k y b(1 + 2k + c 0 /k − )) u1 2bk x k y u 0 , (S76) (δy(t)) 2 BM = k y 2b (1 + τ b k + c 0 ) 2 u 0 g * BM − 1 + 2 τ b k + c 0 − τ b 2 k 2 + c 2 0 u 1 t u 0 + 2bk y 1 + 2 τ b k + c 0 − τ b 2 k 2 + c 2 0 + k x 3 + 2 τ b k + c 0 + τ b 2 k 2 + c 2 0 u 1 bk y k x u 0 , (S77) where g * BM = 1 + (k + c 0 /k − ) 2 . [/formula]
Note that for c 1 = u 1 = 0, the solutions coincide with the solutions for the constant regime. Although it is clear that BM is less noisy than CM for k + c 0 < k − , by comparing the time-dependent terms we find that, in fact, BM is always less noisy than CM. The analytical solutions are plotted in S5 match the numerical simulations with noise. Again, all these calculations were done using Wolfram Mathematica 8, while all the simulations of the stochastic differential equations (Eqs. S59 and S60) were done using the Euler method in MATLAB.
## Further investigations into the accuracy
In this section we provide further explanations for the accuracy of concentration sensing by a single receptor without comparing with the maxmimum-likelihood estimation . In ? we showed results from the master equation for the two regimes k + c 0 < k − and k + c 0 > k − for slow and fast switching of the receptor. Despite its burstiness, the BM receptor turned out more accurate than the CM receptor in the k + c 0 < k − regime for fast switching.
Additional results from the master-equation model. To understand this result better we also implemented an intermediate-modulation (IM) receptor, which has features of both the CM and BM receptors. Like the CM receptor, the IM receptor signals while in the bound (on) state, but instead of a constant rate α of production it produces protein with a rate α ′ so that in each bound interval the same number of molecules are produced irrespective of the interval length, i.e. α ′ τ b = ζ, with ζ the constant burst size of BM. For this to work, the IM receptor would have to know at the time of binding when it will unbind again, in order to choose the correct rate of production. Since the rate of unbinding is a random variable this is generally not possible. Nevertheless, the IM receptor may help to further elucidate our observed trends in accuracy. In practice, we implemented this IM receptor by first simulating a time trace of bound and unbound time intervals with a Gillespie algorithm, allowing us to determine the rate of production as a function of time. Afterwards, the actual protein production and degradation were simulated.
In analogy to ? the results for the IM receptor are shown in S1 , which also shows the results for the CM and BM receptors for comparison in blue and red, respectively. As expected, for slow switching the IM receptor has intermediate accuracy between CM and BM. CM is most accurate as continuous production during the bound intervals is balanced by degradation so the output protein level does not fluctuate excessively. BM is least accurate due to the increased burst size for slow switching. Since signaling by the IM receptor is only burst-like for the short bound intervals but not for the long bound intervals, it is somewhat more accurate than BM. Due to the non-constant rate of production, IM also fluctuates more than CM. This intermediate accuracy is clearly demonstrated by the time traces in the left panels of S2 In the k + c 0 < k − regime for fast switching, the inset of S1C shows that BM is now most accurate and that IM has again intermediate accuracy. While BM steadily produces the same amount of protein at the times of binding, IM produces this amount only during short bound intervals as its rate of production is then high, while during long bound intervals its slow production is buffered by degradation, so its protein level fluctuates more strongly. CM is even worse than IM since, due to its constant rate of production during bound time intervals, it hardly produces any protein during short bound intervals, which leads to drastic drops in protein level, while it produces a lot during long bound intervals due to its constant rate of production.
In contrast, in the k + c 0 > k − regime for fast switching, CM is generally most accurate due to its approximately constant rate of production throughout time, i.e. the receptor is almost always bound and active. IM is less accurate than CM because its rate of protein production is variable due to the variable length in bound intervals, despite the fact that the receptor is mostly bound. Interestingly, IM is even less accurate than BM under these conditions.
Inspecting the examples of time trace in the bottom right panel of S2 the burst sizes of IM can exceed the burst sizes of BM for unusually short bound intervals since production is very high and stochastic, and only on average the same amount of protein is produced during bound intervals than during a burst in BM. During long bound intervals the rate of production is very low. Hence, compared to BM, degradation prevents a net increase in protein level during a bound interval, leading to further variability. A special case is when the burst size ζ is 1. As shown in the inset of ?D, BM can be more accurate than CM. This is because the burst size of BM is minimal and in the master equation the production with minimal rate α in CM is highly stochastic.
As we now discuss, to provide further intuition for the differences in accuracy between the k + c 0 < k − and k + c 0 > k − regimes, we also simulated the variance of the signaling output (and hence the accuracy-determining factor g) directly (see Eq. 5).
Signaling output from ODE model without protein production and degradation. Factor g in Eq. 5 (and Eq. S31) determines the variance of the signaling rate u(t) without invoking any downstream protein production and degradation. For a given time interval ∆t, we can hence simulate u(t) directly. We assess the accuracy of CM, IM, and BM by plotting the histograms of the integrated signaling rate u I (∆t) := ∆t 0 u(t)dt and by determing their variances (cf. derivation of g in . As slow protein production and degradation strongly affect the accuracy of the final protein output for slow switching, this approach mainly helps understand the interesting fast switching case.
We initially assume signaling during bound intervals is deterministic, leading to a linear increase of u with slope α (α ′ ) during a bound time interval for CM (IM) and a step increase by ζ for BM. At each unbound time interval, IM and BM have the same level of signaling output as IM produces the same number of proteins deterministically during each bound interval (ζ). In contrast, the signaling output from CM is generally different since the rate of signaling is always the same for each bound interval but their durations vary. Resulting time traces and variances are shown in S3A and B Figs. left panels, respectively. Specifically, S3A left panels shows clearly that for k + c 0 < k − BM and IM are most accurate with u I (t) increasing almost linearly in time. Since signaling is deterministic, BM and IM are essentially identical, and their variance may only differ due to small differences in signaling during the final bound interval (S3A . This last bound time interval may be interrupted in IM, but for long ∆t this difference is negligible. In contrast, S3B left panels show clearly that for k + c 0 > k − CM is most accurate, as u I (t) is now almost linear in time.
Signaling output from master-equation model without protein production and degradation. Allowing signaling to be stochastic does not change the results for the accuracy significantly. S3A show that for k + c 0 < k − BM is now most accurate and that IM has intermediate accuracy (between BM and CM) due to its variability in signaling in line with S1C show that CM is still most accurate but also that IM is worse than BM in line with S1D Taken together, these additional simulation results confirm our findings of the main text that BM is most accurate for k + c 0 < k − and CM is generally most accurate for k + c 0 > k − .
## Am is more accurate than fm for multiple receptors/ion channels
Here, we provide a more detailed discussion of the accuracy of encoding by multiple receptors, i.e. using AM and FM. To determine whether AM or FM is more accurate in encoding and decoding, we generalize to multiple receptors (or ion channels) (S6 . We assume that AM is obtained by unsynchronized CM receptors (S6A , while FM is obtained by synchronized receptors that individually operate with BM (S6D . Other types of synchronization are also possible with synchronized CM receptors shown in S6B or variable amplitude (S6C in contrast to the data .
To estimate the accuracy, we first consider perfect synchronization and unsynchronization in either modulation scheme. For N unsynchronized (us) receptors, we can express the resulting average and variance of the encoded input by the single-receptor quantities, i.e. u(t) us N = N u(t) and δu(t)δu(t ′ ) us N = N δu(t)δu(t ′ ) 1 . As a result, the relative variance (variance divided by the average-squared) scales with N −1 . In contrast, for N synchronized (s) receptors, the average and variance of the encoded input can be written as u(t) s N = N u(t) and δu(t)δu(t ′ ) s N = N 2 δu(t)δu(t ′ ) 1 , respectively. Hence, the relative variance is now independent of N , so unsynchronized receptors have an N times smaller noise than synchronized receptors. Since N unsynchronized CM receptors lead to AM, we obtain for its relative variance
[formula] δu(t)δu(t ′ ) AM N u(t) AM N 2 = δu(t)δu(t ′ ) CM 1 N u(t) CM 1 2 .(S78) [/formula]
Conversely since N synchronized BM receptors lead to FM, the relative variance of FM is
[formula] δu(t)δu(t ′ ) F M N u(t) F M N 2 = δu(t)δu(t ′ ) BM 1 u(t) BM 1 2 .(S79) [/formula]
For slow dynamics, or fast dynamics with k + c > k − , CM is more accurate than BM. Hence, for N receptors, AM is even more accurate than FM. In contrast, for fast dynamics with k + c < k − , BM is up to twice as accurate as CM (Eq. (??)), and AM is N times more accurate than CM. Consequently, AM becomes more accurate for encoding than FM for more than two receptors (S6E . An exception are two receptors, for which AM and FM can be equally accurate (S6E . Since we generally show that larger signaling noise leads to larger output noise, the same rule emerges for decoding. To extend our results to intermediate levels of synchronization for N > 2 receptors we consider a fraction ρ of synchronized receptors while the remaining fraction (1 − ρ) are unsynchronized, with signaling either by CM or BM (S6E . When comparing CM and BM receptors for the same levels of synchronization ρ, BM receptors can remain more accurate than CM receptors (S6E . However, intermediate levels of synchronization do not strictly represent AM and FM. As shown in S6B,C Figs. synchronized CM receptors lead to pulses of variable duration, while unsynchronized BM receptors lead to highly frequent pulses with potentially variable amplitude.
Taken together, since single cells have thousands of receptors and ion channels, AM is the most accurate modulation scheme. Note that this figure is similar to ? in main text with the addition of IM. Two regimes are shown: k− = 10 k+c0 (α = 100s −1 , γ = 1s −1 , ζ from 1000 to 1) (left column) and k− = 0.1 k+c0 (α = 10s −1 , γ = 1s −1 , ζ from 1000 to 1) (right column). Averages from CM, BM and IM are constrained to be equal, i.e. ζ (BM) = αk −1 − (CM) = α ′ τ b (IM). Variances of CM, BM and IM exhibit two different regimes for fast switching: for k+c0 < k− BM is the most accurate mechanism and CM the worst (inset in C) while for k+c0 > k− CM is generally the most accurate (except for ζ = 1) and IM the worst (inset in D). Third moments show that, for large noise, the probability distributions become asymmetric.
[fig] Figure 1: Experimental evidence for amplitude and frequency modulation. (A and B) Example data showing amplitude modulation from [10]. (A) Single-cell nuclear localization of Msn2 transcription factor in response to H 2 O 2 stress as a function of time. The stimulus profile (input) is a step change applied at t = 0 (inset) which applies to all figure panels. (B) Average time trace for different concentrations of H 2 O 2 stress. (C and D) Example data showing frequency modulation from [15]. (C) Single-cell nuclear localization of Crz1 in response to calcium stress as a function of time, showing bursts of Crz1. (D) The average frequency of bursts against calcium concentration, showing an increased frequency with increased concentration. (Inset) Burst duration distribution for low (blue bars) and high (red bars) concentration. Both histograms are well described by the Gamma distribution h(t) = te −t/τ b , with τ b = 70s (black solid line), demonstrating that pulse duration is independent of calcium concentration. Experimental data in arbitrary units (AU) of fluorescence. [/fig]
[fig] Figure 2: Advantages and disadvantages of amplitude and frequency modulation. AM may be less noisy than FM (A,B), but FM may allow coordinated expression of many genes (C,D) [15, 19]. (A) In AM, low/high stimuli result in low/high levels of transcription factor (TF) inside the nucleus. (B) In AM, different nuclear TF concentrations (blue and red curves) lead to gene expression of proteins A and B (see orange and green promoter functions respectively) with variable ratios (order of dot and square changes). (C) In FM, the stimulus strength only affects the frequency of bursts, not their amplitude. (Inset) Schematic of TF (purple dots) binding promoter P A of gene A (orange) and promoter P B of gene B (green) with different binding strengths. (D) [/fig]
[fig] Figure 4: The two regimes in the linear pathway model based on the master equation. [/fig]
[fig] Figure 5: First three moments of the protein distribution in concentration sensing from the master equation. Averages (A,B), variance (C,D), and skewness (E,F) as a function of the frequency of binding events [/fig]
[fig] S1: Fig. First three moments of the protein distribution in concentration sensing from the master equation. Averages (A,B), variance (C,D), and skewness (E,F) as a function of the frequency of binding events, f = k + c 0 /(1 + k + c 0 /k − ). (Insets) Magnification of small-noise approximation region (fast switching). Analytical results for CM (blue) and numerical results for BM (red) and intermediate modulation IM (green) as function of the frequency of binding events (logarithmic scale). Note that this figure is similar to Fig. 5 in main text with the addition of IM. Two regimes are shown: k − = 10 k + c 0 (α = 100s −1 , γ = 1s −1 , ζ from 1000 to 1) (left column) and k − = 0.1 k + c 0 (α = 10s −1 , γ = 1s −1 , ζ from 1000 to 1) (right column). Averages from CM, BM and IM are constrained to be equal, i.e. ζ (BM) = αk −1 − (CM) = α ′ τ b (IM) [/fig]
[fig] S2: Fig. Examples of time traces of receptor activity and protein copy numbers for different regimes. (Top) Regime [/fig]
[fig] S6: Fig. From CM (BM) to AM (FM) for multiple receptors/ion channels. (A-D) Schematic of receptor activity in time. (A) AM emerges from N unsynchronized receptors or ion channels in CM mode. (B) N synchronized CM receptors lead to a hybrid mechanism with information encoded in the frequency of broad bursts of variable duration. (C) N unsynchronized BM receptors provide a dense series of bursts. For large N , bursts may start overlapping, leading to variable amplitudes. (D) FM emerges from N synchronized receptors in BM mode. (E) Relative variance for a system of 8 receptors with ρN synchronized and (1 − ρ)N unsynchronized receptors, plotted for fast dynamics in the k + c < k − regime (CM in blue and BM in red). Letters refer to panel labels (A-D). Dotted red line indicates uncertainty from FM for comparison. (Inset) Same for a system of two receptors only. [/fig]
[fig] .S3: Examples of time traces of receptor activity and protein copy numbers for different regimes. (Top) Regime k+c0 < k− with k+c = 0.1 k− (α = 100s −1 , γ = 1s −1 ). (Bottom) Regime k+c0 > k− with k+c = 10 k− (α = 10s −1 , γ = 1s −1 ). (Left) Slow switching with ζ = 400. (Right) Fast switching with ζ = 7. Receptor activity r and protein copy numbers n(t) for CM, BM and IM are shown in black, blue, red and green, respectively. Fig. Investigating accuracy based on accumulative signaling (without protein production and degradation). (A) Regime k+c0 < k− with k+c = 0.1 k− (α = 100s −1 , γ = 1s −1 and ζ = 7). (Left) ODE model. (Right) Stochastic protein production during τ b in CM and IM. (Top) Examples of time traces. (Bottom) Histograms of number of proteins produced after 100s with standard deviation in legend based on 1000 simulations. (B) Analogous to (A) but for regime k+c > k− with k+c = 10 k− (α = 100s −1 , γ = 1s −1 and ζ = 7). CM, BM and IM are shown in blue, red and green, respectively. Incoherent feedforward loop: Comparison of analytical results with simulations of the stochastic differential equations. (A) Averages of signaling rate u (left), species y from Eq. (S42) (middle) and species x from (S41) (right) as a function of time. Analytic solutions Eqs. (S32), (S43) and (12) are shown for BM in red, while a (time averaged) time-trace from a stochastic simulation using the Euler method is shown in orange (CM is almost identical and hence is not shown). (B) Corresponding variances as a function of time for k+c0 > k− (k− = 6.7 × 10 5 s −1 , k+c0 = 10 6 s −1 ). Analytic results are shown in blue for CM and in red for BM; average over time (1s) from numerical simulations are shown in light blue for CM and in orange for BM. (C) Corresponding variances as a function of time for k+c0 < k− (k− = 6.7 × 10 6 s −1 , k+c0 = 10 6 s −1 ). Colors same as in (B). Remaining parameters: k+c1 = 10 4 s −2 , kx = 10s −1 and ky = 50s −1 . [/fig]
|
The relationship between temperament, polygenic score for intelligence and cognition: A population‐based study of middle‐aged adults
We investigated whether temperament modifies an association between polygenic intelligence potential and cognitive test performance in midlife. The participants (n = 1647, born between 1962 and 1977) were derived from the Young Finns Study.Temperament was assessed with Temperament and Character Inventory over a 15-year follow-up (assessed with a polygenic score for intelligence. Cognitive performance (visual memory, reaction time, sustained attention, spatial working memory) was assessed with CANTAB in midlife. The PGSI was significantly associated with the overall cognitive performance and performance in visual memory, sustained attention and working memory tests but not reaction time test. Temperament did not correlate with polygenic score for intelligence and did not modify an association between the polygenic score and cognitive performance, either. High persistence was associated with higher visual memory (B = 0.092; FDR-adj. p = 0.007) and low harm avoidance with higher overall cognitive performance, specifically better reaction time (B = À0.102; FDRadj; p = 0.007). The subscales of harm avoidance had different associations with cognitive performance: higher "anticipatory worry," higher "fatigability," and lower "shyness with strangers" were associated with lower cognitive performance, while the role of "fear of uncertainty" was subtest-related. In conclusion, temperament does not help or hinder one from realizing their genetic potential for intelligence. The overall modest relationships between temperament and cognitive performance advise caution if utilizing temperament-related information e.g. in working-life recruitments.Cognitive abilities may be influenced by temperament variables, such as the drive for achievement and anxiety about test performance, but they involve distinct systems of learning and memory.
# | introduction
Cognitive abilities such as working memory, episodic memory and sustained attention, are highly heritable, with the estimates of the proportion of inheritance being around 50% or above and increasing toward late adulthood. [bib_ref] The heritability of general cognitive ability increases linearly from childhood to young..., Haworth [/bib_ref] [bib_ref] The new genetics of intelligence, Plomin [/bib_ref] [bib_ref] Heritability of neuropsychological measures in schizophrenia and nonpsychiatric populations: a systematic review..., Blokland [/bib_ref] [bib_ref] A genome-wide study of common SNPs and CNVs in cognitive performance in..., Need [/bib_ref] [bib_ref] Genetic influences on cognitive function using the Cambridge neuropsychological test automated battery, Singer [/bib_ref] During the last two decades, numerous candidate genes for intelligence have been proposed, but the findings regarding candidate genes have mostly not been replicated. [bib_ref] The impact of genetic research on our understanding of normal cognitive ageing:..., Payton [/bib_ref] [bib_ref] Most reported genetic associations with general intelligence are probably false positives, Chabris [/bib_ref] Recently, several large-scale genome-wide association studies (GWAS) have enabled a more reliable identification of the polygenic architecture of cognitive abilities 2,8 one of them being based on metaanalysis conducted by Savage and coworkers. [bib_ref] Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links..., Savage [/bib_ref] This analysis is built on the firmly established concept, originally produced by [bib_ref] Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links..., Savage [/bib_ref] [bib_ref] general intelligence," objectively determined and measured, Spearman [/bib_ref] that the different aspects of cognitive functioning including verbal and mathematical ability, abstract reasoning, processing speed, executive functioning, spatial reasoning and memory-are considerably captured by a single underlying latent factor, labeled general intelligence or "g." Still, there is substantial variation between different cognitive aspects in their g-loading. [bib_ref] How can general intelligence composites most accurately index psychometric g and what..., Farmer [/bib_ref] The genome-wide meta-analysis by Savage et al., [bib_ref] Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links..., Savage [/bib_ref] combined data from 14 independent cohorts totaling 269,867 participants of European ancestry and 9,295,118 genetic variants associated with cognitive performance in various tests, resulting in the identification of 190 novel loci and 939 novel genes, and replicating previous associations with 15 loci and 77 genes. The resulting polygenic score explained 5.2% of the variance in cognitive performance in four independent samples. [bib_ref] Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links..., Savage [/bib_ref] Even though there is considerable evidence supporting the concept of polygenic scores for cognitive ability, they still explain only a minor share of the total variance of cognitive performance, [bib_ref] The new genetics of intelligence, Plomin [/bib_ref] [bib_ref] Genomic prediction of cognitive traits in childhood and adolescence, Allegrini [/bib_ref] because actual performance is likely to be dependent on complex interactions among groups of genes (i.e., epistasis), [bib_ref] Evolution of genetic networks for human creativity, Zwir [/bib_ref] and on a great variety of factors commonly interacting with genes. Interacting factors include, for example, familial socioeconomic environment, childhood education and environmental cognitive stimulation. [bib_ref] Early childhood cognitive development and parental cognitive stimulation: evidence for reciprocal geneenvironment..., Tucker-Drob [/bib_ref] [bib_ref] Center-based child care and cognitive skills development: importance of timing and household..., Votruba-Drzal [/bib_ref] [bib_ref] Analysis of behavioral traits in the presence of cultural transmission and assortative..., Rice [/bib_ref] For example, an interaction between genetic propensity for intelligence and SES on cognitive performance has been reported both from twin [bib_ref] Large cross-national differences in gene socioeconomic status interaction on intelligence, Tucker-Drob [/bib_ref] and GWAS design studies, [bib_ref] Modification of heritability for educational attainment and fluid intelligence by socioeconomic deprivation..., Rask-Andersen [/bib_ref] even if contradicting findings also exist. [bib_ref] Multivariable GE interplay in the prediction of educational achievement, Allegrini [/bib_ref] Analogously to the interaction found regarding SES, it possible that temperament might interact with genes and as a result help or hinder the realization of one's genetic intelligence potential. Temperament is a set of early emerging, partially heritable dispositions, which are relatively stable over the lifetime and among other things, describe how an individual reacts to novel stimuli. [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] It has been convincingly showed that individuals are differentially susceptible to the psychosocial environment depending on their temperament. [bib_ref] Differences in sensitivity to parenting depending on child temperament: a meta-analysis, Slagt [/bib_ref] [bib_ref] The association between low socioeconomic status and depressive symptoms depends on temperament..., Jokela [/bib_ref] [bib_ref] Nature and nurture in personality, Keltikangas-Järvinen [/bib_ref] [bib_ref] Temperament and character modify risk of drug addiction and influence choice of..., Milivojevic [/bib_ref] Also, one single-gene study has found that temperament may modify an association of genetic factors with academic performance in adulthood. [bib_ref] Novelty seeking as a mediator in relationships between type 4 dopamine receptor..., Keltikangas-Järvinen [/bib_ref] Still, evidence concerning the existence of potential interaction between genes and temperament on cognitive performance is lacking.
The present study was taken with a purpose to examine whether temperament modifies an association between the genetic intelligence potential and performance in cognitive tests in midlife. That is, we investigated whether there are temperament dimensions that impair or promote realizing the genetic intelligence potential. Existence of such, would be an important discovery warranting implications, for example, considering whether individuals with certain temperament would need support to realize their intelligence potential.
Temperament was assessed in terms of Cloninger's psychobiological theory of personality. [bib_ref] A systematic method for clinical description and classification of personality variants: a..., Cloninger [/bib_ref] [bib_ref] A psychobiological model of temperament and character, Cloninger [/bib_ref] According to Cloninger et al., [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] temperament is the "disposition of a person to learn how to behave, react emotionally, and form attachments automatically by associative conditioning." According to [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] in addition to general intelligence, humans have three distinct systems of learning and memory (i.e., associative conditioning, intentionality and self-awareness), which are associated with different components of personality. Associative conditioning (i.e., how we learn to react automatically, including classical and operant conditioning) is the evolutionally first emerged system and forms the basis for temperament. [bib_ref] A psychobiological model of temperament and character, Cloninger [/bib_ref] Further, the associative learning system has been empirically connected with genes having neural functions related to cognitive abilities, including memory and cognitive flexibility. [bib_ref] Uncovering the complex genetics of human temperament, Zwir [/bib_ref] Cloninger's psychobiological model proposes that there are four temperament dimensions: novelty seeking, harm avoidance, reward dependence and persistence. Novelty Seeking refers to activation behavior and the tendency to approach novel stimuli. [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] Harm Avoidance refers to the tendency to inhibitory behavior in the presence of aversive stimuli. Reward dependence refers to the tendency to respond positively and maintain the behavior in the presence of reward signals, while persistence refers to the tendency to maintain the behavior despite the lack of reward. [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] Although the temperament traits are comparatively stable overage, there are some maturational changes: Persistence typically slightly increases and novelty seeking typically decreases over age. [bib_ref] Maturity and change in personality: developmental trends of temperament and character in..., Josefsson [/bib_ref] Temperament might exert its effect in two ways. First, it may have a long-term effect via one's way to react to environmental stimuli: that is, whether one approaches or avoids stimuli that might help to develop cognitive abilities and how one capitalizes on possibilities to learn. [bib_ref] The psychological costs and benefits of being highly persistent: personality profiles distinguish..., Cloninger [/bib_ref] In this way, temperament may precede cognitive performance. Second, temperament may have a situation-related (cross-sectional) effect on performance. [bib_ref] Cloninger's temperament dimensions and threat, stress, and performance appraisals during different challenges..., Ravaja [/bib_ref] That is, it may promote [bib_ref] Personality and intelligence: persistence, not self-directedness, cooperativeness or self-transcendence, is related to..., Mousavi [/bib_ref] [bib_ref] same but different": associations between multiple aspects of self-regulation, cognition, and academic..., Malanchini [/bib_ref] or impair [bib_ref] Neuropsychological assessment anxiety: a systematic review, Dorenkamp [/bib_ref] [bib_ref] Test anxiety effects, predictors, and correlates: a 30-year meta-analytic review, Von Der Embse [/bib_ref] one's test performance as has been widely documented in previous studies. More specifically, temperamental dispositions such as a tendency to be fearful or worried may decrease one's test performance, first, via increased stress and anticipation of failure prior to a cognitive task and, second, via increasing perceived stress and reducing one's ability to concentrate and continue despite fatigue during a task. [bib_ref] Cloninger's temperament dimensions and threat, stress, and performance appraisals during different challenges..., Ravaja [/bib_ref] Previous cross-sectional studies have shown an association of high persistence and low harm avoidance with higher cognitive performance while reported associations regarding novelty seeking or reward dependence with cognitive performance have been inconclusive. [bib_ref] Personality and intelligence: persistence, not self-directedness, cooperativeness or self-transcendence, is related to..., Mousavi [/bib_ref] [bib_ref] On the relationship between autistic traits and executive functioning in a non-clinical..., Maes [/bib_ref] [bib_ref] Personality and cognitive functions in violent offenders -implications of character maturity?, Seidl [/bib_ref] Additionally, several studies have examined the associations of Big Five personality traits with cognitive performance. [bib_ref] Personality predicts cognitive function over 7 years in older persons, Chapman [/bib_ref] [bib_ref] The association between facets of conscientiousness and performance-based and informant-rated cognition, affect,..., Sutin [/bib_ref] [bib_ref] The interplay between personality and cognitive ability across 12 years in middle..., Wettstein [/bib_ref] First, high conscientiousness, which is a trait correlating with high persistence, [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] [bib_ref] Cloninger's psychobiological model of temperament and character and the five-factor model of..., De Fruyt [/bib_ref] is found to associate with higher cognitive performance. [bib_ref] The association between facets of conscientiousness and performance-based and informant-rated cognition, affect,..., Sutin [/bib_ref] [bib_ref] A meta-analysis of the five-factor model of personality and academic performance, Poropat [/bib_ref] [bib_ref] Personality traits prospectively predict verbal fluency in a lifespan sample, Sutin [/bib_ref] [bib_ref] Genetically-mediated associations between measures of childhood character and academic achievement, Tucker-Drob [/bib_ref] Second, high harm avoidance and particularly its subscale of anticipatory worry are known to correlate with high neuroticism [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] [bib_ref] Cloninger's psychobiological model of temperament and character and the five-factor model of..., De Fruyt [/bib_ref] [bib_ref] Harm avoidance and disability in old age, Wilson [/bib_ref] that, in turn, is found to predict lower cognitive performance. [bib_ref] Genetically-mediated associations between measures of childhood character and academic achievement, Tucker-Drob [/bib_ref] However, as far as we know, there is a lack of studies to examine how temperament precedes cognitive performance.
Taken together, we investigated if Cloninger's temperament dimensions are associated with over-or under-realization of polygenic intelligence potential. In addition, we examined both prospective and cross-sectional associations between temperament and test performance; that is, how temperament precedes cognitive performance, and how temperament correlates with cognitive performance crosssectionally. This study utilized the population-based data of "The Cardiovascular Risk in Young Finns Study" (YFS), where the subjects have been followed over 30 years. This makes possible both cross-sectional and prospective study designs. On the basis of previous literature, we hypothesized that high persistence and low harm avoidance are associated with higher cognitive performance. [bib_ref] Personality and intelligence: persistence, not self-directedness, cooperativeness or self-transcendence, is related to..., Mousavi [/bib_ref] [bib_ref] Personality and cognitive functions in violent offenders -implications of character maturity?, Seidl [/bib_ref] We did not set any hypothesis regarding novelty seeking and reward dependence because previous findings on the topic have been inconclusive. Our investigations regarding interactions between temperament and polygenic intelligence potential were exploratory by nature.
## | materials and methods
## | participants
The study participants were members of the YFS. They were selected randomly from six birth cohorts who were living nearby the Finnish universities with medical schools. The sampling was conducted using the Finnish population register of the Social Insurance Institution. The YFS was conducted in accordance with the Declaration of Helsinki, and the ethical committees of all the Finnish universities with medical schools approved the study design at the beginning of the study. A more detailed description of the study population is found elsewhere. [bib_ref] Cohort profile: the cardiovascular risk in young Finns study, Raitakari [/bib_ref] The total sample of the YFS included 3596 participants (ethnic We included all the participants with data available on (i) sex and age, (ii) polygenic score for intelligence (PGSI), (iii) at least one domain of cognitive performance and (iv) all the temperament dimensions in at least one measurement year. For each single temperament scale, all the participants with responses for at least 95% of the items were included. Accordingly, the final data in our analyses included 1647 participants.
## | midlife cognitive performance
The cognitive performance of the participants was assessed with the Cambridge Neuropsychological Test Automated Battery (CANTAB), which is a computerized test measuring several cognitive domains.The full CANTAB test battery includes altogether 24 individual tests and has been shown to have good construct validity and discriminant validity. [bib_ref] Normative data from the CANTAB. I: development of executive function over the..., De Luca [/bib_ref] [bib_ref] Cambridge neuropsychological test automated battery (CANTAB): a factor analytic study of a..., Robbins [/bib_ref] In the current study, we used a test battery with four tests that The Reaction Time test measured reaction time and response accuracy. The Rapid Visual Information Processing test measured sustained visual attention. The Spatial Working Memory test measured spatial working memory and the ability to solve problems using selforganized search strategies.A detailed description of the cognitive performance testing procedure has been reported previously. [bib_ref] Cognitive performance in young adulthood and midlife: relations with age, sex, and..., Rovio [/bib_ref] Each cognitive test produced several outcome variables (e.g., reaction time, number of errors, movement time). A standardized sum variable was calculated for each test by first, transforming each variable into a scale with a mean of 0 and a SD of 1. Then we calculated testwise scores by summing the individual standardized variables within each test and then divided it by the number of variables in that particular test. An overall cognitive performance score was calculated as an average of the testwise sum variables. The calculation of the CANTAB variables is described in more detail elsewhere. [bib_ref] Cognitive performance in young adulthood and midlife: relations with age, sex, and..., Rovio [/bib_ref]
## | polygenic score for intelligence
For each participant, we calculated a polygenic score to reflect their genetic intelligence potential. The genotyping was performed for 2443 samples using a custom build Illumina Human 670 k BeadChip at Welcome Trust Sanger Institute. Genotypes were called using Illuminus clustering algorithm. [bib_ref] A genotype calling algorithm for the illumina bead array platform, Teo [/bib_ref] Genotype imputation was done using Beagle software [bib_ref] A one-penny imputed genome from next-generation reference panels, Browning [/bib_ref] and The Sequencing Initiative Suomi (SISu) as reference data. A polygenic score for intelligence was calculated using LDpred, a Bayesian method that estimates posterior mean causal effect sizes from genome wide association (GWA) study summary statistics by assuming a prior for the genetic architecture and linkage disequilibrium (LD) information from a reference panel [bib_ref] Modeling linkage disequilibrium increases accuracy of polygenic risk scores, Vilhjálmsson [/bib_ref] : an infinitesimal model of causal variants was assumed, and summary statistics from Savage et al., [bib_ref] Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links..., Savage [/bib_ref] GWA study for intelligence were used (https://ctg.cncr.
## | temperament dimensions
Temperament dimensions were measured using the temperament and character inventory (TCI).The TCI includes four temperament dimensions that are novelty seeking, harm avoidance, reward dependence and persistence. The scale of Novelty Seeking (NS) consists of 40 statements (e.g., "I do things spontaneously"), Harm Avoidance (HA) with 35 statements (e.g., "I avoid meeting strangers"), Reward Dependence (RD) with 24 statements (e.g., "I'm strongly moved by sentimental appeals"), and Persistence (PS) with eight statements (e.g., "I often push myself to exhaustion"). All the statements were self-rated with a five-point Likert scale ranging from 1 (completely disagree) to 5 (completely agree).
We calculated the average score of the items for each temperament dimension (four dimensions) at each measurement year. The average scores were standardized with the sample mean and SD. We The construct validity and test-retest reliability of the TCI temperament scales have been shown to be good in previous studies. [bib_ref] A psychobiological model of temperament and character, Cloninger [/bib_ref]
## | statistical analyses
We conducted statistical analyses using the RStudio 1.4.17. We examined attrition by comparing the included (n = 1647) and excluded (n = 1949) participants with regard to study variables (in the comparisons, we included those excluded participants who had data available on each study variable). In attrition analyses, we used independent samples t tests and chi-squared tests. We conducted multivariate linear regression analyses to examine the associations of PGSI and temperament dimensions with cognitive performance. The statistical requirements (e.g., normality, homoscedasticity) for the linear regression analyses were scrutinized graphically and statistically and found to be met appropriately (for further details, please see Supplementary Methods and .
In all analyses, dependent variables included both the overall cognitive performance and the performance in the four cognitive domains of the CANTAB (PAL, RTI, RVP, SWM),given that overall cognitive performance does not capture entirely the variation in different aspects of cognitive performance. The temperament dimensions (NS, HA, RD and PS) were added as the independent variables simultaneously to all analyses. In the cross-sectional analyses, we used temperament dimensions measured in 2012 (i.e., in the same year with cognitive test performance); and in the prospective analyses, we used the average scores of temperament dimensions between measurement years 1997, 2001 and 2007 (i.e., the years preceding cognitive test performance). First, we investigated whether temperament dimensions are associated with cognitive performance when controlling the PGSI. Thus, the temperament dimensions and PGSI were set as independent variables. Next, we investigated whether the temperament dimensions modify the association between the PGSI and cognitive performance. Hence, in that analysis we also included all the twoway interaction effects between PGSI, each temperament dimension and control variables, as has been recommended previously. [bib_ref] Gene environment interaction studies have not properly controlled for potential confounders: the..., Keller [/bib_ref] That is, we examined whether participants scoring high on a certain temperament dimension, such as HA, would have higher cognitive performance if polygenic score for intelligence is high, whereas they would have low cognitive performance if polygenic score for intelligence is low. In all the analyses, control variables included sex, age, and the 10 first ancestrally informative principal components. The principal components were calculated before the polygenic score calculation, and they were estimated using quality-controlled GWAS chip data (the principal components were calculated with Plink software by using Plink's-pca command).
For the variance explained by each independent variable, we calculated the adjusted R squared when separately adding the independent variable in question to the model (including also the covariates and the other independent variables). Because multiple analyses were done, we applied false discovery rate-correction (FDR), [bib_ref] Controlling the false discovery rate: a practical and powerful approach to multiple..., Benjamini [/bib_ref]
# | results
The descriptive statistics of the study variables are presented in . The mean age of the included participants was 42.9 (range 35-50) years at the time of the cognitive performance measurement.
Fifty-six percent of the participants were female.
The attrition analyses showed that the included participants (56% As shown in [fig_ref] Table 2: presents the results of linear regression analyses when predicting cognitive performance by... [/fig_ref] As additional analyses, we reran the models without including the PGSI as a covariate and present the results in [fig_ref] Table 2: presents the results of linear regression analyses when predicting cognitive performance by... [/fig_ref]. Overall, all the significant associations between temperament and cognitive performance remained, and some of the associations seemed to become slightly stronger. , no significant interactions emerged between temperament and PGSI (a more detailed table of the results regarding covariates is provided in .
## As shown in
HA independently explained 0.9% of the reaction time and Persistence independently 0.8% of the visual memory performance. Thus, hypothetically, two individuals would differ by $0.42 SD in their reaction time performance and $ 0.37 SD by visual memory depending on their HA and PS being high versus low (+/À 2 SD).
As additional analyses, we examined whether the subscales of HA were associated with cognitive performance (i.e., the only temperament dimension with significant effects on cognitive performance and with subscales) when entered into the analysis together with the PGSI. The results of regression analysis are shown in . To summarize the findings, all four HA's subscales (viz., low "anticipatory worry," high "fear of uncertainty," low "fatigability," and high "shyness with strangers") had associations with higher cognitive performance. Specifically, low "anticipatory worry" was associated prospectively and cross-sectionally with higher visual memory p = 0.014). We also reran these analyses without PGSI as covariate (see .
# | discussion
Our results indicated that temperament was not related to the suggested genetic background of intelligence potential, and did not modify an association between the polygenic score for intelligence and cognitive performance, either. No significant interaction between the PGSI and temperament dimension survived after the correction for multiple testing. The findings of the current study suggest that a role of temperament in cognitive performance is equal at each level of the polygenic score for intelligence. This means that no temperament dimension seems to give extra advantage or hindrance to an individual's cognitive performance, be their genetic intelligence potential high or low. In other words, temperament does not put individuals on advantaged or disadvantaged position in realizing their potential, in contrast what has been found in regards certain environmental factors (e.g., socioeconomic status) that have been reported modify the realization of genetic intelligence potential. [bib_ref] Large cross-national differences in gene socioeconomic status interaction on intelligence, Tucker-Drob [/bib_ref] [bib_ref] Attained SES as a moderator of adult cognitive performance: testing gene-environment interaction..., Zavala [/bib_ref] Importantly, we see two other potential explanations for the lack of significant interactions.
First, as it is proposed that temperament should be also analyzed as combinations of dimensions, that is, temperament profiles, [bib_ref] The complex genetics and biology of human temperament: a review of traditional..., Cloninger [/bib_ref] disposition to concentrate on the task and to strive to perform at one's best. [bib_ref] The psychological costs and benefits of being highly persistent: personality profiles distinguish..., Cloninger [/bib_ref] [bib_ref] same but different": associations between multiple aspects of self-regulation, cognition, and academic..., Malanchini [/bib_ref] In the current study, an association was found only with performance in visual memory and associative learning. This is, however, in accordance with Cloninger's theory where temperament refers to the differences in associative and nonverbal learning and memory. [bib_ref] A systematic method for clinical description and classification of personality variants: a..., Cloninger [/bib_ref] An association between Persistence and cognitive performance was found only with the 2012 measurement and not with the 1997-2007. In addition to an imperfect stability of temperament across time, this might suggest a more contemporary effect of persistence.
That is, high persistence may be related to better cognitive performance in a test situation (e.g., overcoming challenges in a test situation) instead of having a major role in the long-term development of cognitive performance. A temporary effect of persistence has been previously documented at least in a Finnish experimental study, [bib_ref] Cloninger's temperament dimensions and threat, stress, and performance appraisals during different challenges..., Ravaja [/bib_ref] and in a sample of violent offenders, [bib_ref] Personality and cognitive functions in violent offenders -implications of character maturity?, Seidl [/bib_ref] and adolescents. [bib_ref] Personality and intelligence: persistence, not self-directedness, cooperativeness or self-transcendence, is related to..., Mousavi [/bib_ref] [bib_ref] The psychometrics and validity of the junior temperament and character inventory in..., Moreira [/bib_ref] [bib_ref] The psychobiological model of personality and its association with student approaches to..., Moreira [/bib_ref] Interestingly, however, although persistence may only temporarily promote cognitive test performance, it may prospectively enhance higher educational attainments. [bib_ref] Adulthood temperament and educational attainment: a population-based cohort study, Mullola [/bib_ref] That is, high Persistence may enable long-term utilization of one's cognitive abilities in practice, in order to reach higher educational goals.
We found that low Harm Avoidance was associated with higher cognitive performance both cross-sectionally and prospectively. In addition, these findings are consistent with Cloninger's theory, [bib_ref] A systematic method for clinical description and classification of personality variants: a..., Cloninger [/bib_ref] High "anticipatory worry" (HA1) and high "fatigability" (HA4)
were cross-sectionally associated with and prospectively preceded a lower cognitive performance in different cognitive tests which is plausible as individuals high on these subscales are prone to be anxious, become tired quickly, and give up easier in straining tasks.These tendencies may manifest in a test situation as higher test anxiety that, in turn, associates with lower test performance. [bib_ref] Neuropsychological assessment anxiety: a systematic review, Dorenkamp [/bib_ref] [bib_ref] Test anxiety effects, predictors, and correlates: a 30-year meta-analytic review, Von Der Embse [/bib_ref] [bib_ref] Temperament and intelligence: a psychometric approach to the links between both phenomena, Strelau [/bib_ref] In addition, individuals with high harm Avoidance are prone to have lower performance appraisal and to anticipate failure, 32 that may impair their ability to do their best.
A role of "fear of uncertainty" (HA2) was slightly complicated: it preceded a lower performance in a visual and a working memory test but a higher one in a sustained attention test. Although apparently mutually contradicting, these findings are not in a conflict with previous literature: feelings of uncertainty and acute stress are suggested to be associated with enhanced attention, [bib_ref] Uncertainty and anticipation in anxiety: an integrated neurobiological and psychological perspective, Grupe [/bib_ref] but with lower memory performance, [bib_ref] The effects of acute stress on episodic memory: a meta-analysis and integrative..., Shields [/bib_ref] and altered functioning of some memory-related brain regions such as the hippocampus. [bib_ref] Correlated memory defects and hippocampal dendritic spine loss after acute stress involve..., Chen [/bib_ref] This is in accordance with evolutionary meaningful postulations that high arousal may transfer resources from working memory network to selective attention in order to be able to respond to acutely relevant information. [bib_ref] Arousal-biased competition in perception and memory, Mather [/bib_ref] Our finding that high "shyness with strangers" (HA3) was associated with higher cognitive performance is inconsistent with some previous findings on children, [bib_ref] Shyness and vocabulary: the roles of executive functioning and home environmental stimulation, Blankson [/bib_ref] [bib_ref] A longitudinal perspective on the association between cognition and temperamental shyness, Wolfe [/bib_ref] but congruent, for instance, with some findings in adolescents regarding academic performance. [bib_ref] The psychometrics and validity of the junior temperament and character inventory in..., Moreira [/bib_ref] Further, our finding does not completely disagree with Cloninger's theory,
where "shyness" refers to inhibition in social interactions, not necessarily to cognitive challenges.Finally, we found that novelty seeking and reward dependence played no role in cognitive performance. For reward dependence, previous studies have found a marginal or no effect on cognitive performance, [bib_ref] Personality and intelligence: persistence, not self-directedness, cooperativeness or self-transcendence, is related to..., Mousavi [/bib_ref] The findings on novelty seeking, in turn, are in accordance with two previous studies reporting no association between Novelty Seeking and cognitive performance. [bib_ref] Cloninger's temperament dimensions and threat, stress, and performance appraisals during different challenges..., Ravaja [/bib_ref] [bib_ref] Personality and intelligence: persistence, not self-directedness, cooperativeness or self-transcendence, is related to..., Mousavi [/bib_ref] To the best of our knowledge, this study was the first to examine the role of temperament in the realization of genetic intelligence potential in actual cognitive performance. Our findings are in accordance with studies investigating how big five personality traits precede cognitive performance. [bib_ref] Personality predicts cognitive function over 7 years in older persons, Chapman [/bib_ref] remain quite stable in middle adulthood: strongest (but still rather modest) changes appear to occur in processing speed and verbal abilities, 80,81 but those abilities were not measured in this study.
In conclusion, we found that temperament had a modest association with cognitive performance, but it did not have a role in realization of one's intelligence potential. In addition, some associations between temperament and cognition were contradictory to some general suppositions (e.g., the relationship of high shyness with higher cognitive performance). This suggests that it should be carefully considered if aiming to utilize information about temperament traits in recruitment or academic evaluations. Moreover, the findings together with previous evidence indicate that a same cognitive task may induce different threat perceptions, anticipated performance levels and experienced stress levels in individuals with different temperaments. Taken together, the expression and development of cognitive abilities may be influenced by temperament variables, such as the drive for achievement and anxiety about test performance, but they involve distinct systems of learning and memory.
# Acknowledgments
The Young Finns Study has been financially supported by the Acad-
## Conflict of interest
The author declares that there is no conflict of interest.
# Data availability statement
The data sets presented in this article are not readily available because YFS is an ongoing follow-up study and the datasets are not anonymized, and the GDPR prevents public sharing of the data.
Instead, pseudonymized data sets are possible to share on request, and requires a data-sharing agreement between the parties. Requests to access the datasets should be directed to Katri Räikkönen (katri.
[email protected]) or Niklas Ravaja ([email protected]) for psychological data set, to Terho Lehtimäki ([email protected]) for genetic data set, and to Suvi Rovio ([email protected]) for CANTAB data set.
## Orcid
Aino Saarinen https://orcid.org/0000-0003-4495-8360
[fig] Finns: at the baseline in 1980. For this study, temperament dimensions were assessed four times: in fifth, sixth, seventh and eighth follow-ups of the YFS. That is, in 1997 (participants were 20-35 years old), 2001, 2007, and 2012 (participants were 35-50 years old), enabling prospective examination of the temperament dimensions. Cognitive performance was measured in 2012, and polygenic score for intelligence was calculated after Savage's et al. GWA study 9 on intelligence was published. [/fig]
[table] Table 2: presents the results of linear regression analyses when predicting cognitive performance by PGSI and temperament dimensions. The PGSI was significantly associated with the overall cognitive performance and performance in visual memory, sustained attention and working memory tests but not reaction time test. Our additional analyses showed that the PGSI independently explained 2.3%, 1.5%, 2.0% and 0.8% of the variation in overall cognitive performance, visual memory, sustained attention test and working memory test, respectively. Further, there were no significant associations between PGSI and any of the four temperament dimensions. [/table]
|
Reconstruction of White Matter Tracts via Repeated Deterministic Streamline Tracking – Initial Experience
Diffusion Tensor Imaging (DTI) and fiber tractography are established methods to reconstruct major white matter tracts in the human brain in-vivo. Particularly in the context of neurosurgical procedures, reliable information about the course of fiber bundles is important to minimize postoperative deficits while maximizing the tumor resection volume. Since routinely used deterministic streamline tractography approaches often underestimate the spatial extent of white matter tracts, a novel approach to improve fiber segmentation is presented here, considering clinical time constraints. Therefore, fiber tracking visualization is enhanced with statistical information from multiple tracking applications to determine uncertainty in reconstruction based on clinical DTI data. After initial deterministic fiber tracking and centerline calculation, new seed regions are generated along the result's midline. Tracking is applied to all new seed regions afterwards, varying in number and applied offset. The number of fibers passing each voxel is computed to model different levels of fiber bundle membership. Experimental results using an artificial data set of an anatomical software phantom are presented, using the Dice Similarity Coefficient (DSC) as a measure of segmentation quality. Different parameter combinations were classified to be superior to others providing significantly improved results with DSCs of 81.02%64.12%, 81.32%64.22% and 80.99%63.81% for different levels of added noise in comparison to the deterministic fiber tracking procedure using the two-ROI approach with average DSCs of 65.08%65.31%, 64.73%66.02% and 65.91%66.42%. Whole brain tractography based on the seed volume generated by the calculated seeds delivers average DSCs of 67.12%60.86%, 75.10%60.28% and 72.91%60.15%, original whole brain tractography delivers DSCs of 67.16%, 75.03% and 75.54%, using initial ROIs as combined include regions, which is clearly improved by the repeated fiber tractography method.
# Introduction
Multimodal navigation guidance is a routine tool in neurosurgical operating theaters to achieve best possible resection of the lesion with minimum postoperative morbidity, displaying outlines of the segmented objects in the microscope heads-up display. In addition to mere anatomical magnetic resonance imaging (MRI) data, the multimodality concept includes the visualization of functional brain areas with cortical sites (functional MRI, magneto-encephalography), fiber bundles (diffusion tensor imaging (DTI) based fiber tractography) or metabolic data (single photon emission computed tomography, positron emission tomography, magnetic resonance spectroscopy imaging). To date, the concept of maximum resection whilst preserving neurological functions is not only self-evident for benign lesions, but also for malignancies such as gliomas, as the most common primary brain tumors [bib_ref] The influence of the extent of surgery on the neurological function and..., Vecht [/bib_ref]. Whereas the correlation of their extent of resection (EOR) and patient outcome has been a long-term point of discussion, recent literature favors also radical EOR in surgery of low-grade and high-grade gliomas [bib_ref] The influence of the extent of surgery on the neurological function and..., Vecht [/bib_ref] [bib_ref] Surgical management of intracranial gliomas-does radical resection improve outcome?, Laws [/bib_ref] [bib_ref] Prognostic factors for survival of patients with glioblastoma: recursive partitioning analysis, Lamborn [/bib_ref] [bib_ref] A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of..., Lacroix [/bib_ref] [bib_ref] Volumetric assessment of glioma removal by intraoperative high-field magnetic resonance imaging, Nimsky [/bib_ref]. Another major addition to multimodal navigation is intraoperative MRI (iMRI) followed by an update of the navigation to compensate for the effects of brain shift (brain deformations due to e.g. loss of cerebrospinal fluid, tumor resection, edema) [bib_ref] Comparing 0.2 tesla with 1.5 tesla intraoperative magnetic resonance imaging analysis of..., Nimsky [/bib_ref] [bib_ref] Intraoperative high-field-strength MR imaging: implementation and experience in 200 patients, Nimsky [/bib_ref] [bib_ref] Safety, efficacy, and functionality of high-field strength interventional magnetic resonance imaging for..., Hall [/bib_ref] [bib_ref] Development and implementation of intraoperative magnetic resonance imaging and its neurosurgical applications, Black [/bib_ref]. It was demonstrated that iMRI combined with navigation guidance and an intraoperative update of image data leads to higher rates of EOR and gross total resection rates in glioma surgery with low postoperative morbidity [bib_ref] Intraoperative MRI to guide the resection of primary supratentorial glioblastoma multiforme-a quantitative..., Schneider [/bib_ref] [bib_ref] Impact of intraoperative high-field magnetic resonance imaging guidance on glioma surgery: a..., Hatiboglu [/bib_ref] [bib_ref] Glioma resection in a shared-resource magnetic resonance operating room after optimal image-guided..., Bohinski [/bib_ref] [bib_ref] Quantification of glioma removal by intraoperative high-field magnetic resonance imaging: an update, Kuhnt [/bib_ref].
In order to maximize the extent of tumor volume resection whilst preserving neurological functions, additional image data sets can be used to display more information on the tumor or risk structures such as vessels, subcortical neuronal pathways or eloquent cortical areas. In this paper, we focus on the reconstruction and visualization of subcortical fiber bundles, delivered by DTI and fiber tractography. DTI has first been described by Basser et al. [bib_ref] MR diffusion tensor spectroscopy and imaging, Basser [/bib_ref] in 1994. Upon this finding, DTI based fiber tractography became a popular noninvasive method to estimate the normal course, location, and extent of white matter tracts, as well as displacement or interruption around a tumor or widening of fiber bundles due to edema or tumor in vivo [bib_ref] Measurement of fractional anisotropy using diffusion tensor MRI in supratentorial astrocytic tumors, Beppu [/bib_ref] [bib_ref] White matter fiber tracking in patients with space-occupying lesions of the brain:..., Clark [/bib_ref] [bib_ref] Diffusion-tensor imaging of white matter tracts in patients with cerebral neoplasm, Witwer [/bib_ref] [bib_ref] Brain fiber tracking with clinically feasible diffusion-tensor MR imaging: initial experience, Yamada [/bib_ref].
DTI relies on diffusion weighted imaging (DWI). The fundamental principle of DWI relies on measuring diffusion properties of water molecules in the human brain. Brownian motion of water molecules is random without any preferential direction of movement, which changes in the presence of structures in the area of interest. The generally disordered diffusion process becomes directional in regions with strongly aligned microstructure, e.g. cell membranes, and the myelin sheath. Thereby, the diffusion process of water molecules is strengthened along the aligned microstructural architecture and hindered in the transverse direction. For each DWI acquisition, a diffusion gradient is applied allowing the measurement of the diffusion process within a specified direction [bib_ref] Diffusion tensor imaging: concepts and applications, Bihan [/bib_ref].
DTI uses 2 nd order tensors to describe the diffusion properties within each voxel. Since the positive and symmetric diffusion tensor requires six coefficients, at least six diffusion weighted images using non-collinear diffusion gradients in addition to one non-diffusion weighted image (b0-image) are indispensable. According to the Stejskal-Tanner equation, the coefficients of the tensors can be determined; the principal eigenvector encodes the dominant diffusion direction, corresponding dominant tissue structure and the mean longitudinal direction of axons in major white matter tracts for each volume element [bib_ref] From the diffusion coefficient to the diffusion tensor, Bihan [/bib_ref].
Until now, several DTI based algorithms have been proposed for reconstructing neuronal pathways in the human brain in vivo [bib_ref] In vivo fiber tractography using DT-MRI data, Basser [/bib_ref] [bib_ref] Fiber tracking: principles and strategies -a technical review, Mori [/bib_ref] [bib_ref] Diffusion tensor imaging and axonal tracking in the human brainstem, Stieltjes [/bib_ref] [bib_ref] Reduction of False Positive Valued Area by Combining Probability Maps, Kreher [/bib_ref] [bib_ref] A Bayesian approach for stochastic white matter tractography, Friman [/bib_ref] , generally separated in deterministic and probabilistic methods.
Initial approaches concentrated on deterministic methods [bib_ref] Non-invasive assessment of structural connectivity in white matter by diffusion tensor MRI, Jones [/bib_ref] [bib_ref] Fiber-tractography via diffusion tensor MRI (DT-MRI), Basser [/bib_ref] [bib_ref] Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging, Mori [/bib_ref] [bib_ref] Tracking neuronal fiber pathways in the living human brain, Conturo [/bib_ref] [bib_ref] Regularization of diffusion-based direction maps for the tracking of brain white matter..., Poupon [/bib_ref] [bib_ref] In vivo fiber tractography using DT-MRI data, Basser [/bib_ref]. Based on the assumption that the principle eigenvector, given by the 2 nd order tensor, is correlated with the main direction of the underlying fiber structure [bib_ref] Estimation of the effective self-diffusion tensor from the NMR spin echo, Basser [/bib_ref] [bib_ref] MR diffusion tensor spectroscopy and imaging, Basser [/bib_ref] , a path is iteratively calculated, starting at a defined seed point and following the direction parallel to the principle eigenvector at each voxel.
The most commonly implemented method in neuronavigation systems is the so-called tensor deflection algorithm (TEND) [bib_ref] White matter tractography using diffusion tensor deflection, Lazar [/bib_ref]. In contrast to traditional streamline tractography based on propagating the streamline in the major eigenvector field, TEND uses the whole diffusion tensor to deflect the incoming tract vector [bib_ref] Tensorlines: advenctiondiffusion based propagation through diffusion tensor fields, Weinstein [/bib_ref] towards the major eigenvector direction with limited deflection curvature, resulting in smoother tracts.
Besides these common techniques, many fiber tractography algorithms have been developed. Merhof et al. presented a connectivity based approach using the A* algorithm for path finding between two selected regions, which was particularly evaluated in case of the language pathways [bib_ref] Fast and accurate connectivity analysis between functional regions based on DT-MRI, Merhof [/bib_ref] [bib_ref] Fiber tractography based on diffusion tensor imaging compared with high-angularresolution diffusion imaging..., Kuhnt [/bib_ref]. Additional global tractography approaches were implemented considering the spatial neighborhood for estimation of fiber orientations [bib_ref] A Hough transform global probabilistic approach to multiple-subject diffusion MRI tractography, Aganj [/bib_ref] [bib_ref] A Bayesian framework for global tractography, Jbabdi [/bib_ref] [bib_ref] Accurate anisotropic fast marching for diffusion-based geodesic tractography, Jbabdi [/bib_ref] [bib_ref] Gibbs tracking: a novel approach for the reconstruction of neuronal pathways, Kreher [/bib_ref] [bib_ref] Estimating distributed anatomical connectivity using fast marching methods and diffusion tensor imaging, Parker [/bib_ref] [bib_ref] Global fiber reconstruction becomes practical, Reisert [/bib_ref].
As opposed to deterministic approaches delivering only one fiber reconstruction per seed point, probabilistic algorithms [bib_ref] Estimating distributed anatomical connectivity using fast marching methods and diffusion tensor imaging, Parker [/bib_ref] [bib_ref] Estimation of anatomical connectivity from diffusion tensor data, Koch [/bib_ref] [bib_ref] A framework for a streamline-based probabilistic index of connectivity (PICo) using a..., Parker [/bib_ref] [bib_ref] Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging, Behrens [/bib_ref] consider multiple pathways per seed point and per point along the reconstructed pathways. One probabilistic fiber tracking method has been developed by Kreher et al. [bib_ref] Reduction of False Positive Valued Area by Combining Probability Maps, Kreher [/bib_ref]. For each voxel ascribed to the fiber bundle, a streamline is propagated through the tensor field. The trajectories are chosen depending on random experiments, in contrast to the trajectory calculation within the deterministic approach. Another probabilistic method is presented by Friman et al. [bib_ref] A Bayesian approach for stochastic white matter tractography, Friman [/bib_ref] using a Bayesian approach for fiber tracking. On a global level, the probabilities of a connection between two areas in the brain are estimated. On a local level, the probability density functions concerning the fiber orientation are estimated using the Bayes' theorem.
For clinical purposes, specific white matter tracts have to be selected. After streamline calculation, followed by reducing the resulting fiber sets using include and exclude regions [bib_ref] Fiber Selection from Diffusion Tensor Data based on Boolean Operators, Merhof [/bib_ref] , representative 3D objects are calculated as wrapping hulls. The wrapping process is commonly based on a stepwise calculation of bounding curves, such as convex hulls along the set of streamlines [bib_ref] Isosurface-based generation of hulls encompassing neuronal pathways, Merhof [/bib_ref]. However, the surface is fully dependent on the tracking results and heavily influenced by tracking errors [bib_ref] Visualization Strategies for Major White Matter Tracts for Intraoperative Use, Nimsky [/bib_ref]. One example of surface rendering by wrapping is described by Ding et al.. Another approach, directly calculating a 3D volume, has been presented by Merhof et al. [bib_ref] Directional Volume Growing for the Extraction of White Matter Tracts from Diffusion..., Merhof [/bib_ref]. Starting from a predefined seed region, the volume is spread out directionally, taking the shape of the local tensor into account and determining the direction of the growing process.
Alternatively, segmentation algorithms can be used, dividable into different levels. A first class of segmentation algorithms uses scalar anisotropy measures derived from the tensor data and applies routinely used image segmentation methods [bib_ref] Level set segmentation and modeling of DT-MRI human brain data, Zhukov [/bib_ref] , losing directional information on the underlying structure. More advanced techniques are based on clustering approaches. On a first level, fiber tracts are reconstructed. Subsequent segmentation is performed using pairwise similarity measures, e.g. the Euclidean distance, the ratio of the length for corresponding portions of the fiber to their overall length [bib_ref] Classification and quantification of neuronal fiber pathways using diffusion tensor MRI, Ding [/bib_ref] or by applying normalized cuts after reducing the fibers to a feature vector [bib_ref] Clustering Fiber Tracts Using Normalized Cuts, Brun [/bib_ref]. Since these depend on previous tractography, a third group of segmentation approaches works directly on the tensor data, without extracting fiber pathways. Using metrics on symmetric positive semi-definite tensor fields, such as the Euclidean metric trace between two tensors [bib_ref] Automatic segmentation of thalamic nuclei from diffusion tensor magnetic resonance imaging, Wiegell [/bib_ref] [bib_ref] Tensor Field Segmentation Using Region Based Active Contour Model. Computer Vision -ECCV, Wang [/bib_ref] or the normalized tensor scalar product [bib_ref] A level set method for segmentation of the thalamus and its nuclei..., Jonasson [/bib_ref] , traditional segmentation approaches such as spectral clustering [bib_ref] Automatic segmentation of thalamic nuclei from diffusion tensor magnetic resonance imaging, Wiegell [/bib_ref] [bib_ref] Segmentation of thalamic nuclei from DTI using spectral clustering, Ziyan [/bib_ref] or level set methods [bib_ref] Tensor Field Segmentation Using Region Based Active Contour Model. Computer Vision -ECCV, Wang [/bib_ref] [bib_ref] A level set method for segmentation of the thalamus and its nuclei..., Jonasson [/bib_ref] are applied. Since these methods concentrate more on the segmentation of discrete tensors, rather than on continuous fiber pathways, more sophisticated metrics such as log-Euclidean metric [bib_ref] Log-Euclidean metrics for fast and simple calculus on diffusion tensors, Arsigny [/bib_ref] , information theoretic metrics [bib_ref] DTI Segmentation using an Information Theoretic Tensor Dissimilarity Measure, Wang [/bib_ref] or affine-invariant metrics [bib_ref] A Riemannian approach to diffusion tensor images segmentation, Lenglet [/bib_ref] [bib_ref] Riemannian Geometry for the Statistical Analysis of Diffusion Tensor Data, Fletcher [/bib_ref] [bib_ref] A Riemannian Framework for Tensor Computing, Pennec [/bib_ref] are used. Locally constrained region based methods [bib_ref] Finsler tractography for white matter connectivity analysis of the cingulum bundle, Melonakos [/bib_ref] [bib_ref] A fuzzy, nonparametric segmentation framework for DTI and MRI analysis: with applications..., Awate [/bib_ref] use the minimization of an energy function in a probabilistic framework.
Although fiber bundle directions are often well estimated with commonly used fiber tracking techniques, the actual size of the fiber bundles is frequently underestimated, which causes severe problems when fiber bundle reconstruction is integrated into neurosurgical applications [bib_ref] Fibertracking does not accurately estimate size of fiber bundle in pathological condition:..., Kinoshita [/bib_ref]. This underestimation of the spatial extent and its tendency to concentrate on the tract center, rather than on the tract borders, can be explained in part with partial volume effects. Partial volume effects cause a decrease in anisotropy due to averaging the diffusion and thereby disturb the main diffusion direction at the white matter tract borders [bib_ref] Uncertainty in diffusion tensor based fibre tracking, Hahn [/bib_ref].
To visualize the uncertainty of fiber orientation in combination with data of trajectories, boot strapping methods were introduced [bib_ref] Confidence mapping in diffusion tensor magnetic resonance imaging tractography using a bootstrap..., Jones [/bib_ref] [bib_ref] Non-parametric statistical analysis of diffusion tensor MRI data using the bootstrap method, Pajevic [/bib_ref] using multiple repetitions (e.g. nine) of image acquisition, generating a large amount of data sets (e.g. 5000) and finally applying tractography to all of them, providing visitation maps for estimation of confidence and uncertainty. Newer methods, such as wild bootstrapping [bib_ref] The wild bootstrap to quantify variability in diffusion tensor MRI, Whitcher [/bib_ref] [bib_ref] Using the wild bootstrap to quantify uncertainty in diffusion tensor imaging, Whitcher [/bib_ref] , overcome the prolonged acquisition times for bootstrapping by generating tensor volumes on the basis of tensor fitting and computing the residuals to the fitted model, with similar tractography results like fiber tractography for conventional bootstrapping [bib_ref] Tractography gone wild: probabilistic fibre tracking using the wild bootstrap with diffusion..., Jones [/bib_ref]. Another approach, following a similar idea, has been presented by Hahn et al. [bib_ref] Uncertainty in diffusion tensor based fibre tracking, Hahn [/bib_ref]. In contrast to repetitions of image acquisition, complex Gaussian noise is added, delivering several data sets with variational noise. Fiber tracking is then performed for all data sets and streamlines are accumulated. Thereby, a widening of aggregation of fibers can be observed with the ability to use the same short image acquisition procedure as for conventional analysis in clinical routine. A new method using a combination of probabilistic fiber tractography and tensor clustering has been suggested by Barbieri et al. [bib_ref] DTI segmentation via the combined analysis of connectivity maps and tensor distances, Barbieri [/bib_ref]. The tractogram provided by probabilistic fiber tractography is used as initial fuzzy segmentation mask, which is iteratively updated according to connectivity information from probabilistic fiber tracking and local tensor clustering. Thereby, the approach incorporates the ability to capture fibers deviating from the main tensor diffusion direction (probabilistic fiber tracking) and the more precise delimitation of bundle borders (tensor clustering).
Due to the still remaining lack of certainty of reliability of the reconstruction methods and the underestimation of fiber bundles using currently approved fiber tractography techniques, in clinical routine the wrapping ''hulls'' are extended, resulting in so called safety hulls. Nimsky et al. [bib_ref] Intraoperative visualization of the pyramidal tract by diffusion-tensor-imagingbased fiber tracking, Nimsky [/bib_ref] showed that a distance of 5 mm between the boundary of the originally reconstructed object and second surface (hull) is sufficient for the corticospinal tract to avoid neurological deficits. In case of bad data quality or vicinity of tumors and edema, it may be necessary to enlarge this distance.
In this paper, we focus on the reconstruction of large fiber bundles, such as the corticospinal tract. As already shown by Hattingen et al. [bib_ref] A standardised evaluation of pre-surgical imaging of the corticospinal tract: where to..., Hattingen [/bib_ref] , the deterministic streamline tracking approach significantly depends on the localization of seed regions. Whereas in their study seed region placement in the primary motor areas yields more successful tracking results even with decreased FA values, seed region placement in the cerebral peduncle creates a higher number of fibers tending to be of higher quality. Due to this dependence on seed region placement and to overcome the influence of manual seed region placement, we combine the results of several fiber tracking reconstructions for final reconstruction and visualization. In contrast to the previously described methods based on physically or artificially enlarged data sets, the proposed method deals with a systematic re-seeding for fiber tractography and fiber bundle representation. This approach will be evaluated using software phantoms with modeled anatomical fiber tracts to have ground truth data for comparison.
# Materials and methods
## Data acquisition
To evaluate the new approach on seed region independent fiber tract visualization, a software phantom based on the BrainWeb project [bib_ref] Design and construction of a realistic digital brain phantom, Collins [/bib_ref] was used with given ground truth to compare against, modeling part of the left corticospinal tract [bib_ref] Assessing Fiber Tracking Accuracy via Diffusion Tensor Software Models, Barbieri [/bib_ref]. To model different qualities of data, the phantom data set was additionally Additionally, the presented method was also applied to MRI data of two healthy volunteers Furthermore, the algorithm was applied to MRI data of a 56 years old male patient with a left precentral anaplastic astrocytoma WHO III and a 65 years old female patient with a left temporoparietal glioblastoma multiforme WHO IV. The same protocol as for the volunteers was used for data acquisition.
Informed written consent was obtained from both patients before MRI data acquisition, as part of a prospective study on patients with primary brain tumors. Study approval was given by the local ethics committee of the University of Marburg. Informed consent was also obtained from both volunteers, members of our research group, for testing MRI acquisition schemes (in coordination with the local ethics committee).
# Image analysis
The procedure is structured into a preprocessing step for seed region calculation, a processing unit for fiber tracking, and a post processing step for object generation. The workflow is summarized and illustrated in [fig_ref] Figure 1: Workflow of the repeated fiber tracking approach [/fig_ref]. Step 1: ROI definition. To set up an initial fiber reconstruction of the desired white matter tract, ROIs are placed manually to define seed and target regions.
In case of the corticospinal tract (and the part of the modeled corticospinal tract in phantom data), the seed region is manually drawn by an experience neurosurgeon outlining the cerebral peduncle in the T1-weighted data set. The second seed region is manually drawn, outlining the precentral gyrus.
As for the arcuate fascicle, activation areas from fMRI data acquisition using a word generation paradigm were used to define Broca's and Wernicke's area as seed and target regions. Alternatively, e.g. in case of non-utilizable fMRI results, seed regions were drawn manually outlining the horizontal part of the arcuate fascicle lateral to the corticospinal tract within coronal images (with FA overlay) [bib_ref] Vulnerability of the frontaltemporal connections in temporal lobe epilepsy, Lin [/bib_ref]. The second seed region is drawn manually outlining the descending portion of the arcuate fascicle in the posterior temporal lobe [bib_ref] Fiber density asymmetry of the arcuate fasciculus in relation to functional hemispheric..., Vernooij [/bib_ref].
Step 2: Initial fiber tractography and fiber mask generation. Using the defined first ROI as seed region, an initial fiber tract is reconstructed using deterministic fiber tractography within the medical image processing platform NeuroQLab(as well as tensor calculation on a set of DWI data). The initial reconstruction result is then restricted by the application of the second ROI (target) used as include region.
The resulting fiber bundle is mapped to a binary mask where each voxel scores the gray value of 1 if it is crossed or touched by a fiber and of 0 otherwise.
Step 3: Centerline calculation. The centerline of the reconstructed fiber bundle is calculated according to Klein et al.. For this purpose, the single streamlines are sampled at n points, and each of the n centerline points is calculated as the average of the corresponding streamline points.
Step 4: Seed region calculation. For each sampled point of the centerline, a plane upright to the centerline's local direction, given by two consecutive centerline points is calculated. Within each plane, according to the preprocessing step of a previous publication on boundary calculation [bib_ref] Boundary estimation of fiber bundles derived from diffusion tensor images, Bauer [/bib_ref] , equally distributed rays are sent out, and the rays are sampled at equally spaced points each. The first point (contour point) of each ray outside of the masked tract volume defines the initial fiber tract outlines. The contour of the resulting generated seed region for the plane is then defined by spline approximation of contour points found for the rays (SCALING 0). Since the deterministic fiber tractography often underestimates the spatial extent of the tract, seed regions with SCALING 0 are enlarged. Therefore, the contour points are moved outwards along the ray according to the SCALING level, i.e. for SCALING 1 the contour points are moved 1 mm outwards.
Step 5: Repeated fiber tractography. The main part relates to the application of the deterministic fiber tracking for all calculated seed regions using the initial seed-and include region as alternative include ROIs. Since the fiber bundle center is well estimated with routinely used algorithms, a combined use of the initial ROIs as include regions would not provide much benefit. The alternative use of the regions allows capturing also areas close to boundaries or branches. After finishing all tracking procedures, each fiber tracking result is assembled into a binary image, as it is also done for the initial tracking result. Finally, all binary masks for the calculated seed regions are combined into one mask including gray values from 0 to n, where n is the number of seed regions.
Step 6: Combination and visualization. For visualization, several objects can be generated by applying a threshold to the final mask image to model the level of fiber bundle membership (FBM) of each voxel. Therefore, different levels are defined. The voxels ascribed to the kernel of the fiber bundle are described by an FBM of 100% to 90% (FBM 90), i.e., 90% or more reconstructions of the fiber bundle include the volume element. In analogy, further objects with lower FBM levels (e.g., FBM 80 = 80% or more reconstructions are included) can be created and displayed as a 3D overlay on the fiber bundle kernel.
For our implementation and evaluation, the medical image prototyping platform MeVisLab (www.mevislab.de) and the neuroimaging prototype NeuroQLabwere used. Additional modules and scripting routines were implemented in C++ and Python, respectively. Evaluations were performed on an Intel Core i7-2600K CPU, 3.4 GHz, 16 GB RAM, Windows 7 Professional, SP1. To evaluate the software phantoms with known courses and extents of the modeled fiber bundles, the Dice Similarity Coefficient (DSC) [bib_ref] Statistical validation of image segmentation quality based on a spatial overlap index, Zou [/bib_ref] is used. The DSC is a widely used measure in medical imaging studies to quantify the degree of overlap between two segmented objects A and B, in this case the reference segmentation (ground truth) and the algorithmically computed segmentation result. Since a DSC of 1 describes a perfect match of both segmentations, a DSC of 0 indicates that there is no overlap.
The independent variables ''number of seed regions (SEEDS)'', ''seed region scaling (SCALING)'', ''fiber bundle membership (FBM)'' and ''image noise (NOISE)'' are varied systematically according to Table 1. In total, 1440 different parameterizations were evaluated.
For a comparison with standard procedures, deterministic fiber tractography was performed using the two-ROI-approach, with the first ROI as seed region and the second ROI as target region. In addition, original whole brain tractography, using the whole brain as seed volume, and a variant of whole brain tractography, using the whole set of calculated seed regions as seed volume, were performed. Tractography results were in both cases restricted using the initial seed ROI and include ROI as combined include ROIs (AND).
Statistical analysis was performed in IBM SPSS Statistics 20 for Windows (SPSS Inc., Chicago, Illinois). The level of significance is set to p,0.05. Analysis of variance homogeneities was performed applying the Levene test for each independent variable (NOISE, SEEDS, SCALING, FBM). According to the test results, univariate ANOVA with the post-hoc Tukey-HSD-Test or the Games-Howell-Tests were performed. Comparison of repeated fiber tractography with the deterministic fiber tractography methods was performed using the Wilcoxon-Mann-Whitney-Utest.
Further comparison with probabilistic tractography using the volunteer and patient data was performed within FSL (SMRIB Software Library v. 5.0) [bib_ref] Bayesian analysis of neuroimaging data in FSL, Woolrich [/bib_ref] [bib_ref] Advances in functional and structural MR image analysis and implementation as, Smith [/bib_ref] using the Diffusion Toolbox FDT [bib_ref] Characterization and propagation of uncertainty in diffusion-weighted MR imaging, Behrens [/bib_ref] [bib_ref] Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?, Behrens [/bib_ref] and parameter settings as presented in [bib_ref] Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?, Behrens [/bib_ref].
# Results
## Results of standard deterministic fiber tracking
The standard deterministic fiber tracking was applied to three types of phantom data according to the image noise.
The two-ROI-approach took only a few seconds each, as well as the variant of the whole brain tractography and original whole brain tractography.
The two-ROI approach resulted in a DSC of 65.08% 65.31% (range: 56.99%-73.50%) for noise level 0, for noise level 1 in a DSC of 64.73% 66.02% (range: 54.85%-74.15%). The tracking procedure applied to the image with noise level 2 resulted in a DSC of 65.91% 66.42% (range: 57.72%-74.64%).
The adapted version of whole brain tractography also led to an underestimation of the tract volume with a DSC of 67.12% 60.86% (range: 65.38%-67.59%) for noise level 0, a DSC of 75.10% 60.28% (range: 74.65%-75.33%) for noise level 1 and a DSC of 72.91% 60.15% (range: 72.61%-72.99%) for noise level 2. Original whole brain tractography led to an underestimation of the tract volume with a DSC of 67.16% for noise level 0, a DSC of 75.03% for noise level 1 and a DSC of 75.54% for noise level 2.
## Analysis of repeated fiber tracking results
Univariate ANOVA according to the Levene test was performed for the variables NOISE and SCALING with the post-hoc Tukey-HSD-Test. For SEEDS and FBM, Games-Howell-Tests were applied.
The time of segmentation via repeated tracking procedure differed according to the number of seed regions used for repeated tractography, reaching from a few seconds (less seed regions) to about 3 minutes for the largest number of seed regions.
## Impact of image noise (noise)
According to the grouping variable NOISE, three subgroups were classified with n = 480 samples for each subgroup.
For [fig_ref] Table 2: Dice Similarity Coefficient [/fig_ref]. According to univariate ANOVA, the image noise does not significantly affect the quality of fiber bundle segmentation (with the DSC as measure of quality) using the presented method (p = 0.794).
## Impact of seed region scaling (scaling)
According to the grouping variable SCALING, eight subgroups were classified with n = 185 samples each. The repeated fiber tracking procedure delivered mean DSCs ranging from 67.41% (group A) to 70.32% (group C), with maximal DSCs of 82.03% to 85.68% (see [fig_ref] Table 3: Dice Similarity Coefficient [/fig_ref].
Significant differences between subgroups related to SCALING were found using the univariate ANOVA (p = 0.008). Post-hoc analysis with the Tukey-HSD-test showed significant differences between group A and group C (p = 0.035), with significantly improved results for group C. No significant differences were found for all other subgroups. Analysis of homogenous subgroups according to Tukey-HSD delivered two homogeneous subgroups 1 (SCALING 0, 1, 3, 4, 5) and 2 (SCALING 1, 2, 3, 4, 5).
## Impact of number of seed regions (seeds)
According to the grouping variable SEEDS, six subgroups were classified each with n = 240 samples each.
The presented procedure resulted in mean DSCs ranging from 65.23% (group A) to 70.41% (group F), with maximal DSCs of 74.64% to 85.68% (see .
The impact of the independent variable SEEDS was evaluated using the Games-Howell-Test. Highly significant differences (p = 0.000) were found between subgroups A and C, A and E, A and F, A and G, A and H with subgroup A providing significantly worse results. Less significant differences were also found between subgroups A and B (p = 0.038) and subgroups A and D (p = 0.023) also giving significantly worse results for subgroup A in comparison to the other groups. All other pair wise comparisons did not deliver significant differences.
## Impact of membership variable (fbm)
According to the grouping variable FBM, ten subgroups were classified with n = 144 samples each.
Mean DSCs ranging from 52.57% to 79.92% were reached with the repeated fiber tracking approach according to FBM classification; with maximal DSCs in the range of 62.32% to 85.68% (see [fig_ref] Table 5: Dice Similarity Coefficient [/fig_ref].
According to Games-Howell-Tests no significant differences were found for subgroups C, D, E (C-D: p = 0.598, D-E: p = 0.054, C-E: p = 0.997), for FBM-subgroups B and F (p = 0.963), subgroups A and G (p = 0.472) and A and H (p = 0.914). All other possible pair wise combinations of FBMsubgroups gave highly significant differences (p = 0.000). Considering Dunnett's-C-Test, subgroups C, D, E are significantly better (according to DSC) than all other groups; subgroups B and F are superior to subgroups A, G, H, I, J. Subgroup G provided better results than subgroups H, I and J, and subgroups A and H gave better results than groups I and J. Finally, subgroup I provided better results than subgroup J.
The results of the repeated tracking approach (see [fig_ref] Figure 2: 3D view of repeated fiber tracking results [/fig_ref] do not differ significantly according to the evaluated image noise. The scaling used for seed region generation shows significantly better results for a scaling of 2 mm in contrast to no scaling, providing two homogeneous subgroups with scaling from 1 mm to 5 mm and one subgroup with 0 mm offset and 2 mm to 5 mm offset. Since 2 mm scaling should be preferred over 0 mm scaling, parameterizations of the first subgroup should be considered for good fiber segmentation results. The analysis of the number of seed regions shows significant differences only between the use of . Application of repeated fiber tracking on healthy volunteer data: corticospinal tract. Comparison (from left to right) of the results from standard deterministic fiber tracking, variant of whole brain fiber tracking, whole brain fiber tracking, probabilistic fiber tractography, unfiltered results of the repeated fiber tracking method and filtered results of the repeated fiber tracking method with a fiber bundle membership of 50% using a seed region scaling of 2 mm and 128 generated seed regions. doi:10.1371/journal.pone.0063082.g007
2 seed regions and all other possible evaluated numbers of seed regions; more than 2 seed regions resulted in improved segmentation results. The evaluation of the fiber bundle membership variable delivers significantly different results, best for 30%, 40% and 50% FBM, see [fig_ref] Figure 3: Influence of fiber bundle membership [/fig_ref]. . Application of repeated fiber tracking on healthy volunteer data: corticospinal tract. Comparison (from left to right) of the results from standard deterministic fiber tracking, variant of whole brain fiber tracking, whole brain fiber tracking, probabilistic fiber tractography, unfiltered results of the repeated fiber tracking method and filtered results of the repeated fiber tracking method with a fiber bundle membership of 50% using a seed region scaling of 2 mm and 128 generated seed regions. doi:10.1371/journal.pone.0063082.g008 . Application of repeated fiber tracking on patient data with a left precentral glioma: corticospinal tract. Comparison (from left to right) of the results from standard deterministic fiber tracking, variant of whole brain fiber tracking, whole brain fiber tracking, probabilistic fiber tractography, unfiltered results of the repeated fiber tracking method and filtered results of the repeated fiber tracking method with a fiber bundle membership of 50% using a seed region scaling of 2 mm and 128 generated seed regions, tumor segmented manually in red. doi:10.1371/journal.pone.0063082.g009
In total, best results were obtained for a scaling of 1 mm to 5 mm, more than 2 seed regions and a fiber bundle membership of 30-50%.
Analysis of this subset of acquired data (n = 105 for each class of NOISE) results in higher values of DSC in correlation to the standard deterministic tracking procedure with only one seed region resulting in DSCs of 65.08% 65.31% (NOISE 0), 64.73% 66.02% (NOISE 1) and 65.91% 66.42% (NOISE 2). The named subset results in mean DSC values of 81.02% 64.13% (NOSISE 0), 81.32% 64.22% (NOISE 1) and 80.99% 63.81% (NOISE 2), see [fig_ref] Table 6: Dice Similarity Coefficients [/fig_ref] and [fig_ref] Figure 4: Comparison of standard fiber tracking and repeated fiber tracking [/fig_ref]. Comparing the subgroups of repeated fiber tracking with deterministic fiber tracking procedures used by default, the standard procedures differed significantly for all noise levels (p = 0.000) according to Wilcoxon-Mann-Whitney-U-test from the achieved results of the presented method. Therefore, the presented approach can be considered as a promising technique to improve the quality of fiber bundle segmentation significantly, see With the resulting information about suitable parameterization to improve fiber tractography results and the noise independent behavior, the presented method was applied to healthy volunteer's data and to two patient data sets with intracerebral gliomas. The corticospinal tract was reconstructed for the leg area on the left and right side in case of volunteer data and on the left side in case of the patient data set including the left precentral glioma. Besides, the arcuate fascicle was reconstructed in case of both volunteer data sets and in case of the patient data set including the left temporo-parietal glioma. For repeated fiber tractography, a seed region scaling of 2 mm, 128 generated seed regions and a fiber bundle membership of 30% to 50% was used.
A comparison of the repeated fiber tractography approach (without and with application of FBM), the two-ROI-approach, the whole brain tractography approach, its variant and additionally probabilistic fiber tractography (connectivity analysis) is presented in and 8 for part of the corticospinal tract encoding motor function for the lower extremities, in two healthy volunteers. Furthermore, tractography was applied to a patient data set, with a precentral anaplastic astrocytoma WHO III. Results are shown in . Further comparison is presented for the arcuate fascicle in [fig_ref] Figure 1: Workflow of the repeated fiber tracking approach [/fig_ref] in case of healthy volunteers and in case of a patient data set with a temporo-parietal glioblastoma multiforme WHO IV. Results are shown in [fig_ref] Figure 1: Workflow of the repeated fiber tracking approach [/fig_ref].
# Discussion
In all cases, application of the repeated fiber tractography approach led to widened fiber bundle segmentation in comparison to results of the two-ROI-approach and improved results of the whole brain fiber tracking approach in concordance with previous tests on the software phantom data. Probabilistic fiber tractography (within FSL) reconstructs only a subset of the corticospinal tract similar to the other methods. Probabilistic fiber tractography thereby took approximately 6 hours (mainly for preprocessing with BEDPOSTX resulting in a distribution of diffusion parameters for each voxel). Two-ROI-approach and whole brain fiber tractography took only a few seconds, the repeated fiber tracking approximately 3 minutes.
For application in the clinical routine, the use of fiber tractography requires short data acquisition times due to patient compliance, workflow efficiency, and also fast reconstruction techniques for segmentation and especially intraoperative update of functional information. Since DTI data acquisition is possible with short acquisition times in the range of 4-5 minutes on a 3T MRI system and approximately 8 minutes on a 1.5T MRI system (with the previous described setting), the fiber tractography algorithms based on DTI data sets have been tested extensively and deterministic streamline fiber tractography is available in most [fig_ref] Figure 1: Workflow of the repeated fiber tracking approach [/fig_ref]. Application of repeated fiber tracking on healthy volunteer data: arcuate fascicle. Comparison (from left to right) of the results from standard deterministic fiber tracking, variant of whole brain fiber tracking, whole brain fiber tracking, probabilistic fiber tractography, unfiltered results of the repeated fiber tracking method and filtered results of the repeated fiber tracking method with a fiber bundle membership of 50% using a seed region scaling of 2 mm and 128 generated seed regions. doi:10.1371/journal.pone.0063082.g010 [fig_ref] Figure 1: Workflow of the repeated fiber tracking approach [/fig_ref]. Application of repeated fiber tracking on healthy volunteer data: arcuate fascicle. Comparison (from left to right) of the results from standard deterministic fiber tracking, variant of whole brain fiber tracking, whole brain fiber tracking, probabilistic fiber tractography, unfiltered results of the repeated fiber tracking method and filtered results of the repeated fiber tracking method with a fiber bundle membership of 50% using a seed region scaling of 2 mm and 128 generated seed regions. doi:10.1371/journal.pone.0063082.g011 surgical planning stations. Intensive clinical evaluation has been performed for the corticospinal tract to estimate accuracy of fiber tracking, still longing for more accuracy, and thereby for increased tumor volume reduction [bib_ref] Fibertracking does not accurately estimate size of fiber bundle in pathological condition:..., Kinoshita [/bib_ref] [bib_ref] Uncertainty in diffusion tensor based fibre tracking, Hahn [/bib_ref] [bib_ref] Intraoperative visualization of the pyramidal tract by diffusion-tensor-imagingbased fiber tracking, Nimsky [/bib_ref].
The presented method initially uses the clinically widespread used tensor deflection algorithm based on DTI, combining the result of different automatically initiated applications, similar to previously published methods like the application of bootstrapping [bib_ref] Confidence mapping in diffusion tensor magnetic resonance imaging tractography using a bootstrap..., Jones [/bib_ref] or wild bootstrapping [bib_ref] Using the wild bootstrap to quantify uncertainty in diffusion tensor imaging, Whitcher [/bib_ref] or the application of variational noise for uncertainty estimation [bib_ref] Uncertainty in diffusion tensor based fibre tracking, Hahn [/bib_ref]. In contrast to the bootstrapping technique, the method uses only one acquired data set and thereby keeps the time requirements of clinical usage. In contrast to wild bootstrapping with a need for a model to fit to and calculate residuals, repeated fiber tracking can easily be adapted to other models and model-free approaches for fiber reconstruction [bib_ref] Tractography gone wild: probabilistic fibre tracking using the wild bootstrap with diffusion..., Jones [/bib_ref]. Additionally, no changes of image quality have to be performed for repeated application of fiber tractography. All approaches mentioned above concentrate on the fiber boundary, with repeated measurement using different data sets, noise varied measurements, or re-seeding using the basic data set.
In contrast to probabilistic fiber tractography methods, that deliver a connectivity map, direct thresholding can be applied for fiber bundle segmentation based on the accumulated fiber tracking results. In case of probabilistic tractography, the connectivity score is distance dependent, requiring an additional re-seeding for final fiber bundle segmentation and application in neurosurgical practice, increasing the processing time.
Up to now, as DTI is still the routinely applied model, fiber tractography algorithms based on the 2 nd order tensor model are used in clinical practice. However, there are many challenging topics arising due to the use of DTI. With its restricted 2 nd order tensor model, assuming Gaussian distributed diffusion, voxels with multi-fiber populations cannot be represented correctly, resulting in an inability to resolve crossing or kissing fibers. Even fanning fiber tracts are hard to resolve with this assumption [bib_ref] Deterministic and probabilistic tractography based on complex fibre orientation distributions, Descoteaux [/bib_ref] , for example making it difficult to reconstruct the whole corticospinal tract on the basis of DTI and streamline tractography. Since approximately one third of voxels in the brain contain more than one fiber population [bib_ref] Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?, Behrens [/bib_ref] , different approaches for description of diffusion functions (e.g., different model, model-free approaches) have to be found. Advanced techniques like high angular resolution diffusion imaging (HARDI)or Q-Ball imaging (QBI) [bib_ref] Q-ball imaging, Tuch [/bib_ref] offer opportunities for the application of advanced diffusion functions to overcome the drawbacks of the restricted tensor model provided by DTI. Up to now, these techniques are not widely used in clinical applications due to their long acquisition times, high hardware-performance requirements and processing times. Acquisition times for scanning protocols using 128 diffusion encoding gradients are up to 20 minutes on a 3T MRI system, and nearly 40 minutes on a 1.5T MRI system, which are most commonly available. Thus, until now these techniques are of lower feasibility in clinical practice. Nevertheless, they will be increasingly suitable for clinical evaluation and clinical use.
Due to the modular structure of the repeated fiber tracking framework, the used diffusion model ''diffusion tensor'' as well as the basic deterministic streamline tractography algorithm can easily be replaced by other techniques resulting in a streamline representation. Approaches resulting in a 3D volume representation of the fiber bundle, like the volume growing approach [bib_ref] Directional Volume Growing for the Extraction of White Matter Tracts from Diffusion..., Merhof [/bib_ref] , need an alternative preprocessing step for centerline calculation, skeletonizing the volume to its centerline.
In the future, alternative approaches will be included for repeated fiber tracking, also based on advanced imaging techniques, if image acquisition and processing could be performed within adequate times (e.g. HARDI with compressed sensing techniques) for example resolving the fanning of the corticospinal tract, or white matter tracts that can hardly be visualized in clinical routine.
For theoretical purposes, time consuming HARDI techniques and advanced fiber tractography algorithms are suitable, however not efficiently applicable in clinical routine.
Our study group currently works on tractography on the basis of currently published theoretical approaches using compressed sensing techniques for reconstruction of HARDI signals using sparse data [bib_ref] Fast and accurate reconstruction of HARDI data using compressed sensing, Michailovich [/bib_ref] [bib_ref] Spatially regularized compressed sensing for high angular resolution diffusion imaging, Michailovich [/bib_ref] in comparison to traditional HARDI acquisition. Initial experience on the reconstruction of the language pathways using HARDI signals with compressed sensing techniques in comparison to DTI related reconstruction based on the same DTI data sets (30 gradient diffusion encoding directions), delivered promising results [bib_ref] Fiber tractography based on diffusion tensor imaging compared with high-angularresolution diffusion imaging..., Kuhnt [/bib_ref]. Until now, HARDI using compressed sensing is not implemented on a commercially available software platform. Thus, integration of these data into neuronavigation systems remains a future perspective. Finally, we aim at the integration in the repeated fiber tractography framework.
The presented approach concentrates on the commonly reconstructed part of the corticospinal tract, as large white matter tract in the human brain, which has been extensively examined and evaluated for clinical purposes. Since the repeated fiber tracking approach delivers promising results for better estimation of the spatial extent of the tract, also in the vicinity of intraaxial tumors , where anatomy is distorted and diffusion patterns might be changed, it will also be adapted to other relevant white matter tracts. A first impression is given by the reconstruction of the arcuate fascicle in consistence with results obtained from the corticospinal tract. Furthermore, this approach [fig_ref] Figure 1: Workflow of the repeated fiber tracking approach [/fig_ref]. Application of repeated fiber tracking on patient data with a left temporo-parietal glioma: arcuate fascicle. Comparison (from left to right) of the results from standard deterministic fiber tracking, variant of whole brain fiber tracking, whole brain fiber tracking, probabilistic fiber tractography, unfiltered results of the repeated fiber tracking method and filtered results of the repeated fiber tracking method with a fiber bundle membership of 50% using a seed region scaling of 2 mm and 128 generated seed regions, tumor segmented manually in red. doi:10.1371/journal.pone.0063082.g012 will be tested for other large white matter tracts such as the optic radiation on basis of the developing advanced methods and evaluated clinically.
# Conclusions
We presented a new fast approach for the reconstruction of fiber bundles in diffusion weighted imaging data using automatically calculated reconstruction along initial tracking results for the desired white matter tract. Evaluation on an anatomical software phantom shows that the presented method is able to improve the segmentation quality of white matter tracts with adequate processing times significantly in contrast to the standard tensor deflection (two-ROI-approach and whole brain tractography approaches), which is routinely used in most clinical settings.
[fig] Figure 1: Workflow of the repeated fiber tracking approach. doi:10.1371/journal.pone.0063082.g001 [/fig]
[fig] Figure 2: 3D view of repeated fiber tracking results. 3D view of results obtained from the repeated tracking approach for an anatomical software phantom with modeled left corticospinal tract, color-coded with green areas indicating regions covered by a large number of fiber tracking applications (high fiber bundle membership) and blue areas indicating regions that are covered by only few fiber tracking results (low fiber bundle membership). doi:10.1371/journal.pone.0063082.g002 [/fig]
[fig] Figure 3: Influence of fiber bundle membership (FBM). Overview of different levels of FBM (left side) with additional overlay of ground truth data in red (right side): FBM 0% (first row), FBM 30% (second row), FBM 50% (third row) and FBM 100% (forth row). doi:10.1371/journal.pone.0063082.g003 [/fig]
[fig] Figure 4: Comparison of standard fiber tracking and repeated fiber tracking. Comparison of results according to the Dice Similarity coefficient (DSC) derived by standard deterministic fiber tracking (in blue) and results of the repeated fiber tracking approach (green) within the group of significantly best parameter settings (n = 315), subdivided by grouping variable NOISE (n = 105). doi:10.1371/journal.pone.0063082.g004 [/fig]
[fig] Figure 5: Comparison of two-ROI-approach, whole brain tractography and repeated fiber tracking (example). Comparison of fiber tracking results in blue achieved by the two-ROI-approach (row 2), traditional whole brain tractography (row 3), variant of whole brain tractography (row 4) and the repeated tracking approach (row 5) in axial, sagittal and coronal view within the anatomical phantom data set and underlying modeled ground truth (row 1) in green. doi:10.1371/journal.pone.0063082.g005 [/fig]
[fig] Figure 6: Visualization of results of repeated fiber tracking. Color-coded results of repeated fiber tracking procedure in sagittal, axial and coronal view with summed up fiber tract reconstructions derived from 128 seed region with light areas showing regions of high consensus between the tracking results (classification according to FBM), additional overlay of modeled corticospinal tract as ground truth in red on the right side. doi:10.1371/journal.pone.0063082.g006 [/fig]
[table] Table 1: Parameterization of independent variables. [/table]
[table] Table 2: Dice Similarity Coefficient (DSC) according to group variable NOISE. [/table]
[table] Table 3: Dice Similarity Coefficient (DSC) according to group variable SCALING. [/table]
[table] Table 5: Dice Similarity Coefficient (DSC) according to group variable FBM. [/table]
[table] 27: years old female and 30 years old male), acquired on a 3T MRI System (Tim Trio, Siemens, Erlangen, Germany) at the University of Marburg including a T1weighted 3D image (3D-Magnetization Prepared Rapid Gradient Echo (MPRAGE): repetition time (TR) 1900 ms, echo time (TE) 2.26 ms, field of view (FoV) 256 mm, matrix 2566256, slice thickness 1 mm, 176 slices, sagittal), a diffusion weighted image data set using single shot echo planar imaging (TR 7800 ms, TE 90 ms, FoV 256 mm, matrix 1286128, slice thickness 2 mm, numbers of excitation 1, b = 1000s/mm 2 , 30 non-collinear diffusion-encoding gradients, voxel size of 26262 mm 3 ) and a functional MRI data set using a word generation task (TR 2000 ms, TE 30 ms, FoV 230 mm, matrix 64664, slice thickness 3.6 mm, voxel size of 3.663.663.6 mm 3 ). [/table]
[table] Table 6: Dice Similarity Coefficients (DSC) for best parameterizations of the repeated fiber tracking approach. [/table]
|
ADAR1 function affects HPV replication and is associated to recurrent human papillomavirus-induced dysplasia in HIV coinfected individuals
Supplementary Figure 1HPV16 E7 pSTAT1 ADAR1 SiHa HaCaT RIG-I MDA5 GAPDH Supplementary Figure 1. Protein expression profile of players in the innate immune response between SiHa and HaCaT cell lines. Western blot of lysates from SiHa (HPV+) and HaCaT (HPV-) cell lines shows higher protein expression of RIG-I and phosphorylated STAT1 in SiHa (HPV16+) compared to HaCaT (HPV16-) cell line. Expression of MDA5, another cytoplasmic RNA sensor, and ADAR1 remained similar between SiHa and HaCaT. HPV16 E7 protein is only present in SiHa cell line, as expected. A representative western blot is shown.
Upregulation of IFN expression in ADAR1 knockdown cells was only seen in Huh 7 cells that harbor a functional RIG-I. Only slight changes in IFN production were observed when RLR, MDA5 and RIG-I, were downregulated in Huh7 cells, whereas no change was seen in the signaling-incompetent RIG-I Huh7.5 cells. Moreover, ADAR1 knockdown was able to significantly increase RLR expression in Huh7 cells, further confirming that ADAR1 function modulates IFN expression through RLR signaling pathway. Gene expression was assessed by quantitative PCR and normalized to GAPDH expression. Data represents mean ± SD of at least 3 independent experiments and is normalized to Mock-transfected cells. siNT, non-targeting siRNA; *p<0.05; **p<0.005; ns, not-significant.
[fig] Supplementary figure 3: Supplementary figure 2. ADAR1 knockdown negatively regulates expression of IFN type I and RLR/MAVS signaling pathway mediators in SiHa cell line. (A) Effective downregulation of ADAR1 by two different siRNA in SiHa cell line. Relative mRNA expression of ADAR1 was measured by quantitative PCR and normalized to GAPDH expression. (B) Relative expression of INFB1A, CXCL10 and IRF7 in siADAR1 SiHa cells. mRNA expression was measured by quantitative PCR and normalized to GAPDH expression. INFB1A, CXCL10 and IRF7 gene expression was upregulated in siADAR1, whereas expression level in Mock-transfected or siNT did not change in SiHa cell line. Data represents mean ± SD of 2 independent experiments and is normalized to Mock-transfected SiHa cells. **p<0.005; ***p<0.0005. ADAR1 knockdown induces expression of IFN type I in RIG-I competent cells. ADAR1 expression was downregulated in Huh7 cells (A) and its derivative cell line Huh-7.5 (B), whose genome encodes a signaling-incompetent endogenous RIG-I (J. Virol. 2005;79:2689-2699). [/fig]
[fig] figure 4 figure 5: Poly(I:C) induces innate immune activation and enhances expression of HPV16 E1 and E7 genes in HPV+ cell line. Relative mRNA expression of (A) interferon stimulated genes (IFNB1, DDX58, IFIH1 IRF7, CXCL10 and ISG15) and (B) HPV16 E1 and HPV16 E7 genes, 16h post-transfection with 2μg of poly(I:C) per 1,25 x10 5 SiHa cells. Gene expression was assessed by quantitative PCR and normalized to GAPDH expression. Data represents mean ± SD of at least 3 independent experiments and is normalized to Mock-transfected cells.*p<0.05; **p<0.005; ***p<0.0005; ****p<0.0001. Gene expression profile of ADAR1, RIG-I and MDA5 knockdowns and ADAR1-RIG-I and ADAR1-MDA5 double knockdowns. (A) Significant downregulation of ADAR1, RIG-I (DDX58) and MDA5 (IFIH1) in all conditions. (B) Significant increase in gene expression of interferon stimulated genes (IFNB1, CXCL10, IRF7 and ISG15) in siRNA-ADAR1 knockdown cells, enhanced IFNB1 in siADAR1-siRIG-I double knockdown and increase CXCL10 expression in siADAR1-siRIG-I and siADAR1-siMDA5 double knockdowns. (C) Significant increase in gene expression of HPV16 E1 and HPV16 E7 only in siRNA-ADAR1 knockdown. 64h post-transfection with 50 pmol of siRNA, or 25 pmol of each siRNA for double knockdown, per 1,25 x10 5 SiHa cells. Gene expression was assessed by quantitative PCR and normalized to GAPDH expression. Data represents mean ± SD of at least 3 independent experiments and is normalized to Mock-transfected cells. *p<0.05; **p<0.005; ***p<0.0005; ****p<0.0001. [/fig]
|
An Atypical Presentation of Severe Dissecting Aortic Aneurysm: A Case Report and Literature Review
# Introduction
Aortic dissection is an uncommon (2.6-3.5 per 100,000 person-years), but morbid condition, with mortality rates, reported around 20-30%. The most common risk factors for dissecting aortic aneurysms are uncontrolled hypertension, atherosclerosis, dyslipidemia, and smoking. Less common risk factors include connective tissue disorders like Marfan syndrome, vasculitis, and traumatic causes. Patients are usually 60-80-year-old men (mean 63y). Women typically present later (mean 67y). It typically presents severely and catastrophically, with chest pain and hemodynamic instability being the most common onset symptoms. This case report demonstrates an atypical presentation that closely mimics that of decompensated congestive heart failure. However, the patient was ultimately diagnosed with a severe thoracic aortic dissection secondary to uncontrolled hypertension, with a paradoxically preserved cardiac efficiency.
## Case presentation
A 55-year-old male patient with no past medical history presented to the ED with complaints of shortness of breath, with exertion, then later at rest, progressive bilateral leg swelling, and generalized fatigue for the preceding two weeks. He endorsed orthopnea and nocturnal paroxysmal dyspnea but denied headache, dizziness, syncope, chest pain, cough, abdominal pain, nausea, vomiting, diarrhoea, constipation, claudication, varicose veins, and limb numbness.
On presentation, the patient was afebrile, body temperature 98.7, tachycardic and hypertensive, heart rate 103 bpm, BP 169/76 mmHg, respiration rate (RR) 20/min, O2 saturation 98% breathing ambient air. Physical examination revealed faint bilateral pulmonary crackles over the lower lung fields, in addition to a 3/6 harsh diastolic murmur auscultated over the left sternal border and apex, with a non-displaced point of maximal apical impulse. It was also noted that the patient had moderate to severe bilateral lower limb pitting edema up to the knees. The remainder of his examination was non-pertinent. His brain natriuretic peptide (BNP) was 15 pg/ml (<100 pg/ml). His lipid panel revealed a cholesterol of 160 mg/dL (120-200 mg/dL), high-density lipoproteins (HDL) 28 mg/dL (29-71 mg/dL), low-density lipoproteins (LDL) 115 mg/dL (<100 mg/dL), and triacyl glycerides 87 mg/dL (40-199 mg/dL). His initial Troponin I was 0.04 ng/ml (<0.5 ng/ml).
A trans-thoracic echocardiogram was obtained, and it revealed the patient's left ventricular ejection fraction to be 50-55%, with hypertensive cardiomyopathy and a globally decreased left ventricular systolic function.
It also revealed a mild mitral valve regurgitant jet, and a mildly dilated pulmonary artery and an elevated pulmonary artery systolic pressure. But the most important finding was a moderate to severe aortic regurgitation, with Type I dissection of the ascending and descending aorta, and the aortic arch (figures 1-2).
## Figure 1: echocardiography demonstrating aortic regurgitation figure 2: ct angiogram of the chest demonstrating aortic double lumen
Upon discovering these findings, a contrast CT angiogram of the chest and an aortogram was obtained, which in turn revealed severe aortic dissection extending from the aortic root all the way down to the iliac bifurcation (figures 3-6). The patient was immediately transferred to the Intensive Care Unit, and started on furosemide for volume status optimization, and started on a labetalol drip for decreasing the blood pressure and ventricular contraction velocity to alleviate the shearing force on the aortic wall (anti-impulse therapy), with a target heart rate of less than 60 bpm and blood pressure less than 120/89 mmHg.
After stabilization, the patient was transferred to another facility and underwent a successful aneurysm surgical repair and aortic valve replacement. He was discharged on continuous anti-impulse therapy for outpatient follow up.
# Discussion
As stated before, aortic dissection is a life-threatening condition that usually presents acutely and with catastrophic outcomes. Mortality rates vary according to patients' age, gender, and stage at presentation. However, according to a whole nation study conducted on 153 patients by Melvinsdottir et al., around 17% died before hospital arrival. In contrast, the risk of death for patients who arrived alive at a hospital was 21.4% within 24h and 45.2% at 30 days. The study also shows that with proper surgical or medical management, depending on the condition, the 30-day mortality rate improves with every year, and the fiveyear survival rate has been increasing steadily.
Aortic dissection can be classified according to a presentation of "acute" and "chronic", depending on the duration of symptoms before a presentation. Less than two weeks between the onset of symptoms and presentation falls into the acute category, and beyond that window, it is labelled a chronic condition. This classification is of significance given that acute onsets are much more liable to complications related to the extension of the dissection and involvement of major aortic branches like the carotids, left subclavian, or even coronaries.
Perhaps a more relevant classification regarding morbidity and mortality rates would be the Stanford classification. According to this classification, a Type A dissection involves the ascending aorta, and a Type B dissection does not. Unfortunately, Type A dissections are almost twice as common as Type B. Moreover, Type A dissection management is generally surgical, while Type B is standardly managed medically.
Mortality in Type A cases managed surgically ranges between 18-22%, depending on postoperative complications. In comparison, mortality among Type A patients managed medically is 57%, a value that naturally explains the inclination to manage Type A surgically. Type B, on the other hand, shows mortality benefit when managed medically, as mortality rates among medically treated patients are around 12% within the first 30 days, while patient's who are managed with surgical or endovascular intervention experienced a mortality rate of 15-21%. A study by Pape et al. in 2015 proposes a mortality rate of 30% in patients with Type B aneurysm who were treated surgically compared to a rate of 10% who were managed medically.
In our case, the most common presenting symptom for a dissecting aneurysm is sharp, severe chest pain, with occasional radiation to the back in >90% of patients, while painless dissections account for less than 6.5% of cases. Signs like hemodynamic instability and pulse and blood pressure disparity between upper limbs, which our patient did not have, present in around 33% of cases, correlate poorly with mortality and complication rates.
Aortic regurgitation is also a common sign in proximal dissections, occurring in 50% of cases. Acute decompensated congestive heart failure is a known complication of severe aortic dissections. However, it typically coexists with syncope, chest pain, abdominal pain, and malperfusion symptoms to vital organs and distal peripheries.
Our patient had acute decompensated heart failure symptoms but none of the other symptoms that usually accompany it. An echocardiography revealed an LVEF of 50-55%, effectively ruling out reduced ejection fraction heart failure. Furthermore, his lab work did not reveal any organ affection, creatinine, and liver enzymes within normal limits.
The most likely etiology behind the development of our patient's condition is undiagnosed, uncontrolled hypertension. He was hypertensive on presentation and stated that he had never seen a physician in his life before. So our patient falls into the narrow fraction of cases with acute, severe dissecting aortic aneurysm with almost none of the typical presenting symptoms. And especially regarding the severity of the condition and the dissection involving the entire aorta from root to bifurcation, his quality of life before presentation and survival are remarkable.
# Conclusions
Severe aortic dissection is a severe and often life-threatening condition, and clinicians are often comfortable with its diagnosis and management, given how drastically it typically presents. However, in this case, we demonstrate a severe case of aortic dissection that presented with almost none of the typical markers of the condition, especially the severity. We hope this case report would raise the medical community's awareness regarding the rare subtle cases that could be missed or misdiagnosed because of atypical presentations.
# Additional information disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work. |
Gravimetric sensors operating at 1.1 GHz based on inclined c-axis ZnO grown on textured Al electrodes
c and * 55 c are the piezoelectrically stiffened elastic constants of the rotated crystal and ρ is the density of ZnO and the value depends on the inclination angle, Y with respect to the surface normal, and ρ is the density. The equations can be used to determine the ratio of the frequencies of the longitudinal to shear modes.The ZnO c-axis inclination angle determined by XRD are shown in 3D inFig. S1. The intensities of the peaks decrease as T S increases from 25 °C to 200 °C. At T S = 300 °C the intensity rises again as the surface roughness of the Al bottom electrode reduces to 14.2 nm. : XRD 3D pole figures for ZnO films deposited on Al sputtered at different T S starting from 25 °C to 300 °C. When T S = 100 °C the c-axis inclination angle of the ZnO has a narrower distribution than at higher temperatures.
The intensity cross-sectional plots along Φ for different T S are shown in [fig_ref] Figure S2: XRD intensity at the position where the maximum intensities occur, with the... [/fig_ref] to determine the FWHM of Ψ and its approximate value. showing that the c-axis inclination angle increases as T S increases but the distribution also increases.
Quantitative data extracted from the XRD pole figures are tabulated in [fig_ref] Table I: Quantitative XRD data at different T S and the Ψ value at... [/fig_ref] , and demonstrate the decrease of the intensity as the surface roughness increases. In addition the grain distribution has a significant impact on the FWHM of Ψ, which increases from 33° to 40° as T S rises, thereby indicating the increased dispersion and hence inhomogeneity of the c-axis inclination angles.
## Fitting with mbvd model
The resonator electrical admittance spectra are fitted with the mBVD model and the values are tabulated in [fig_ref] Table I: Quantitative XRD data at different T S and the Ψ value at... [/fig_ref]. Using the values derived from the mBVD model in [fig_ref] Table I: Quantitative XRD data at different T S and the Ψ value at... [/fig_ref]
[formula] T S (°C) R m (Ω) R s (Ω) R 0 (Ω) L m (μH) C 0 (pF) C m (pF) k 2 eff (%) Q [/formula]
## Measurement in liquid
The electrical admittance spectra of the device without any liquid and with ethanol and water dropped on the active area are shown in . The Q r at resonance without any load is 159, and decreases to 92 in ethanol and to 68 in water. The viscosity (η) and density (ρ) data for different EtOH-water compositions are obtained from Khattab et al. [bib_ref] Density, viscosity, and surface tension of water+ethanol mixtures from 293 to 323K, Khattab [/bib_ref] The values for the compositions and for the product (η ·ρ) 0.5 are listed in the [fig_ref] Table I: Quantitative XRD data at different T S and the Ψ value at... [/fig_ref]. The temperature variation of the room where the measurement is carried out and a typical frequency shift plot with time is shown in . In addition the measurements performed to demonstrate the sensing principles last less than 30 mins, and the temperature variations are therefore not influent on the frequency shifts observed. : Frequency shift plot and temperature variation with time of a non-functionalized resonator to illustrate that the temperature of the room changes by less than ±0.5 °C over long periods.
[fig] Figure S2: XRD intensity at the position where the maximum intensities occur, with the different values of T S [/fig]
[fig] Figure S3: Effect of water and ethanol on the thickness shear mode resonance showing a reduction in the Q r from 159 to 92 in ethanol and 65 in water. Dotted black line corresponds to EtOH, dashed red line to water and solid blue line to device without liquid. [/fig]
[table] Table I: Quantitative XRD data at different T S and the Ψ value at the maximum intensity. [/table]
|
Scalp cooling for reducing alopecia in gynecology oncology patients treated with dose-dense chemotherapy: A pilot project
A B S T R A C TObjective: Determine the efficacy of scalp cooling for the prevention of chemotherapy-induced alopecia in gynecology oncology patients.Methods: This prospective pilot study included patients diagnosed with a gynecological malignancy that received DigniCap™ scalp cooling. Patients were divided into two groups based on chemotherapy regimen: Carboplatin with area under the curve (AUC) 5-6 every three weeks and (1) conventional Paclitaxel 175 mg/m 2 every three weeks or (2) Paclitaxel 80 mg/m 2 weekly. A 1-10 visual analogue scale (1 no hair loss, 10 -complete hair loss) was used to assess degree of hair loss by patients themselves and by a certified dermatologist using photographs. Changes in quality of life and body image were measured using the European Organization for Research and Treatment of Cancer quality of life questionnaire version 3 (EORTC QLQ-C30) and the Body Image Scale (BIS) for cancer patients. Results: Hair preservation occurred with use of a scalp cooling device for patients receiving weekly Paclitaxel (n = 20), but not conventional every three weeks Paclitaxel (n = 8). Ten of 15 patients (66.7%) in the dose-dense group lost less than 50% of their hair based on self-assessment and 14 of 16 (87.5%) based on dermatologist assessment. No patient in this group acquired a cranial prosthesis (wig). There was no difference between groups in terms of quality of life (QoL) and BIS scores. Conclusion: Scalp cooling may allow for hair preservation in gynecology oncology patients receiving Carboplatin AUC 5-6 and weekly Paclitaxel 80 mg/m 2 combination chemotherapy.
# Introduction
Chemotherapy-induced alopecia is a transient but troubling side effect of chemotherapy regimens, influenced by the agent, dose, and scheduling. The psychological effects of hair loss include negative effects on body image, sexuality, self-esteem, and quality of life [bib_ref] Symptom, symptom experiences, and symptom distress encountered by women with breast cancer..., Boehmke [/bib_ref] [bib_ref] The meaning of quality of life in patients being treated for advanced..., Luoma [/bib_ref] [bib_ref] Living in it, living with it, and moving on: dimensions of meaning..., Richer [/bib_ref]. Alopecia has been described as one of the most distressing side effects of chemotherapy [bib_ref] Changing patient perceptions of the side effects of cancer chemotherapy, Carelle [/bib_ref] [bib_ref] Effect of peri-operative chemotherapy on the quality of life of patients with..., Kiebert [/bib_ref] [bib_ref] Psychological sequelae and alopecia among women with cancer, Mcgarvey [/bib_ref] , with effects persisting after patient's hair returned [bib_ref] Embodying identity in chemotherapy-induced alopecia, Koszalinski [/bib_ref] [bib_ref] Cancer and stigma: experience of patients with chemotherapy-induced alopecia, Rosman [/bib_ref].
Scalp hypothermia is becoming increasingly popular in preventing chemotherapy-induced alopecia, with newer systems such as DigniCap™ using helmets with a continuously supplied coolant to maintain a constant temperature. The mechanism of action is vasoconstriction at the level of the scalp to decrease the dose of chemotherapy agent reaching hair follicles [bib_ref] Frontal subcutaneous blood flow, and epi-and subcutaneous temperatures during scalp cooling in..., Bulow [/bib_ref] [bib_ref] Prevention of chemotherapy-induced hair loss by scalp cooling, Grevelman [/bib_ref] [bib_ref] Chemotherapy-induced alopecia: advice and support for hair loss, Roe [/bib_ref] [bib_ref] Effects of temperature and doxorubicin exposure on keratinocyte damage in vitro, Janssen [/bib_ref]. In addition, a reduction in keratinocyte basal metabolic rate makes the hair follicles less reactive to chemotherapy [bib_ref] Prevention of chemotherapy-induced hair loss by scalp cooling, Grevelman [/bib_ref].
Recent systematic reviews and meta-analyses of published randomized control trials show a beneficial effect of scalp hypothermia in preventing hair loss [bib_ref] Scalp Hypothermia for Preventing Alopecia During Chemotherapy. A Systematic Review and Meta-Analysis..., Rugo [/bib_ref] [bib_ref] Scalp hypothermia as a preventative measure for chemotherapyinduced alopecia: a review of..., Shah [/bib_ref] [bib_ref] The scalp cooling therapy for hair loss in breast cancer patients undergoing..., Wang [/bib_ref] [bib_ref] Interventions for Preventing Chemotherapy-Induced Alopecia: A Systematic Review and Network Meta-analysis of..., Zhou [/bib_ref]. These studies used varied chemotherapy regiments, with few including Taxanes and none specifically focusing on the gynecology oncology population. As such, the current prospective pilot study aimed to determine the efficacy of the DigniCap™ scalp cooling system in preventing chemotherapyinduced alopecia for gynecology oncology patients, as well as to observe the effect on quality of life and body image perceived by patients. This is of particular interest since the National Comprehensive Cancer Network (NCCN) recommends considering scalp cooling for patients with ovarian, fallopian tube, or peritoneal cancer in their February 2021 guideline [bib_ref] Ovarian Cancer, Version 2.2020, NCCN Clinical Practice Guidelines in Oncology, Armstrong [/bib_ref].
# Methods
This prospective pilot study was conducted at the division of gynecologic oncology at the Segal Cancer Center of the Jewish General Hospital, a tertiary care hospital in Montreal, Québec. The study is in accordance with the declaration of Helsinki and was approved by the Institutional Review Board with annual reviews. All patients provided informed consent for participation.
## Patient population and data collection
Gynecological oncology patients scheduled to receive chemotherapy were enrolled in the study. Patients were invited to participate if a spot was available for scalp cooling at the time when they were starting chemotherapy, which was limited by the fact that only one machine was available two days per week. Each machine could accommodate two patients at the time. Exclusion criteria were age younger than 18, previous chemotherapy-induced alopecia, scalp conditions that preclude the use of a cold cap, or discontinuing treatment after less than two cycles. The patients were divided in two groups with different timelines of recruitment. Group 1 (January to June 2015) included patients receiving Carboplatin AUC 5-6 and Paclitaxel 175 mg/m 2 every three weeks, with assessments at every treatment. Group 2 (July 2016 to November 2018) included patients receiving Carboplatin AUC 5-6 and weekly Paclitaxel 80 mg/m 2 . At our centre, the dose-dense weekly Paclitaxel was offered liberally to patients as previous assessments showed this regimen to be similar to the standard every three weeks regimen for ovarian and endometrial cancers [bib_ref] Dose dense carboplatin paclitaxel improves progression free survival in patients with endometrial..., Kogan [/bib_ref] [bib_ref] Carboplatin plus paclitaxel weekly dose-dense chemotherapy for high-grade ovarian cancer: A re-evaluation, Kessous [/bib_ref]. Their assessments occurred at different time points:
(1) start of treatment, (2) cycle 3, and (3) cycle 6, with a 3-week before or after flexibility. Baseline data characteristics collected included age, site of malignancy, date of surgery, use of neoadjuvant therapy, and date, type, and number of chemotherapy cycles completed. This data along with the questionnaire responses were collected using Microsoft Excel.
## Scalp cooling
Prior to premedication infusion at each chemotherapy session, a trained volunteer operated the scalp hypothermia machine and helped patients wet their scalps and place the cold cap. Once cooled to 3 - C for 30 min, patients received infusions of their chemotherapy while the scalp cooling device maintained the scalp temperature. Following treatment completion, the machine was turned off and the cold cap was removed 15 min after. Pain during scalp cooling was not assessed. The patients were asked to not dye or cut their hair, to avoid other damaging hair products or blow drying, and to wash their hair at least 48 h after treatment and only once per week. It is important to note that while at our institution the scalp cooling device represented a donation, the cost of one DigniCap™ machine in Canada is 60 000CAD$ and the cost is estimated at 1500-2000US$ per chemotherapy treatment per patient in the United States.
## Alopecia assessment
A visual analogue scale (VAS), with 1 being no hair loss and 10 being complete hair loss, was used to assess degree of hair loss by patients themselves and by a certified dermatologist. The dermatologist used the four photographs obtained for patients at each assessment (front, back, left-side, right-side) for scoring.
## Quality of life
Patient health-related quality of life was evaluated using the European Organization for Research and Treatment of Cancer quality of life questionnaire (EORTC QLQ-C30 version 3) [bib_ref] The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life..., Aaronson [/bib_ref]. This internationally validated 30-item multidimensional questionnaire measures health-related quality of life by generating 15 subscales across three domains: global health status, functional scales, and symptom scales. For the global health status and functional scales, a higher score indicates a better quality of life. For symptom scales, a lower score is preferred. Scores were calculated as per the EORTC QLQ score manual.
## Body image
Body image was evaluated by using the body image scale (BIS) for cancer patients created by [bib_ref] A body image scale for use with cancer patients, Hopwood [/bib_ref] , which includes 10 Likert-items and is scored out of 30, with higher scores indicating a more negative self-image. . Visual analogue scales (VAS) as reported by patient and by dermatologist for each treatment group, with a higher score representing a greater degree of alopecia. Group 1 Represents patients receiving Carboplatin AUC 5-6 and Paclitaxel 175 mg/m 2 every three weeks, While Group 2 represents patients receiving Carboplatin AUC 5-6 and weekly Paclitaxel 80 mg/m 2 as a dose dense.
# Results
## Patient characteristics
A total of 28 patients participated in the pilot project, 8 in Group 1 and 20 in Group 2. The patients in Group 1 discontinued scalp cooling treatment after only two sessions due to significant alopecia. All patient demographics, characteristics, and treatment information are collected in [fig_ref] Table 1: Patient demographics divided into group 1 [/fig_ref]. Four patients were excluded from group 2: two patients received less than two chemotherapy cycles, one patient had experienced chemotherapy-induced alopecia previously, one patient could not wear the cap as biggest size available was too small.
## Alopecia
The visual analogue scale (VAS) scores as perceived by patients and assessed by dermatologists are illustrated in and [fig_ref] Table 1: Patient demographics divided into group 1 [/fig_ref]. For group 1, the assessment for the third cycle was completed by only 4 patients and only 5 patients had photos taken and assessed by the dermatologist, as the others discontinued the scalp cooling treatment. For group 2, the number of patients completing the selfassessment was 14, 18, and 15 for cycles 1, 3, and 6 respectively, while the number of patients having photos taken and assessed by the dermatologist was 14, 17, 16 for cycles 1, 3, and 6 respectively. Within group 2, no patient reported alopecia in the VAS 8-10 range, 3 of 15 patients reported VAS of 6-7, 4 of 15 patients reported 4-5, and 8 of 15 patients reported 2-3 as reported in [fig_ref] Table 1: Patient demographics divided into group 1 [/fig_ref]. For the dermatologists' assessment, 2 of 16 had a VAS of 5-6, all others had VAS score less than 5 within the same group (see [fig_ref] Figure 2: Back-view for one of the patients in Group 2 [/fig_ref].
## Health-related qol and body image scale
The means and standard deviations were calculated for the 15 subscales of the EORTC QLQ-C30 and for the 10 items of BIS are illustrated in [fig_ref] Table 2: European Organization for Research and Treatment of Cancer Quality of Life Questionnaire... [/fig_ref] , with no statistical significant difference between the groups.
# Discussion
To our knowledge, this pilot project represents the first study of scalp cooling in a gynecology oncology population. The first group treated with chemotherapy every three weeks had remarkable alopecia after the second treatment and none of them continued scalp cooling at the third treatment, suggesting the lack of efficacy in this treatment group. As such, the discussion will focus on group 2 receiving dose dense chemotherapy. For group 2, the scalp cooling showed promising results, with 12 of 15 patients reporting less than 50% of hair loss and none of the patients acquiring a cranial prosthesis. Studies have found scalp cooling to be efficient in preventing chemotherapy-induced alopecia [bib_ref] Scalp Hypothermia for Preventing Alopecia During Chemotherapy. A Systematic Review and Meta-Analysis..., Rugo [/bib_ref] [bib_ref] Scalp hypothermia as a preventative measure for chemotherapyinduced alopecia: a review of..., Shah [/bib_ref] [bib_ref] The scalp cooling therapy for hair loss in breast cancer patients undergoing..., Wang [/bib_ref] [bib_ref] Interventions for Preventing Chemotherapy-Induced Alopecia: A Systematic Review and Network Meta-analysis of..., Zhou [/bib_ref] [bib_ref] Sensor-controlled scalp cooling to prevent chemotherapyinduced alopecia in female cancer patients, Fehr [/bib_ref] [bib_ref] COOLHAIR: a prospective randomized trial to investigate the efficacy and tolerability of..., Smetanay [/bib_ref] [bib_ref] Scalp cooling with adjuvant/neoadjuvant chemotherapy for breast cancer and the risk of..., Rugo [/bib_ref] [bib_ref] Effect of a Scalp Cooling Device on Alopecia in Women Undergoing Chemotherapy..., Nangia [/bib_ref]. A 2018 meta-analysis found a relative risk reduction of 43% [bib_ref] Scalp Hypothermia for Preventing Alopecia During Chemotherapy. A Systematic Review and Meta-Analysis..., Rugo [/bib_ref]. However, the studies included mainly breast cancer patients with varying chemotherapy regimens [bib_ref] Effect of a Scalp Cooling Device on Alopecia in Women Undergoing Chemotherapy..., Nangia [/bib_ref] [bib_ref] Impact of alopecia and scalp cooling on the well-being of breast cancer..., Van Den Hurk [/bib_ref]. Whereas all patients in our first group discontinued scalp therapy, none of the patients in group 2 discontinued cooling. This could be explained by less aggressive alopecia induced with the weekly regimen [bib_ref] Impact of chemotherapy regimen and sequence on the effectiveness of scalp cooling..., Villarreal-Garza [/bib_ref] , which allowed for better efficiency of the scalp cooling treatment. In addition, the cap was well tolerated and none of the patients reported complaints with respect to the cooling system. An increased risk of scalp metastasis was a voiced literature concern which was not observed in our patient population and not supported in a recent meta-analysis .
For QoL and BIS measures, the there was no difference between groups despite improvement of alopecia for group 2. These findings are consistent with a recent systematic review showing that scalp cooling is not consistently associated with significant quality of life improvements as assessed by the EORTC QLQ-C30 [bib_ref] The effect of scalp cooling on CIA-related quality of life in breast..., Marks [/bib_ref]. Although reducing hair loss can potentially improve perceived quality of life, it may not compensate for the chemotherapy-related toxicity impact.
The limitations of the current study included its small sample size with lack of statistical power. This was due to the additional cost of the scalp cooling system, with only one machine available in the treatment institution for two patients at a time. In addition, the scalp cooling portion of the treatment leads to an additional hour on the chemotherapy chair, as such only two days per week were allocated for this project due to limitations in nursing resources. The project was halted by the COVID Pandemic and the completion of questionnaires and photographic documentation was not performed by all patients. The VAS scale has an intrinsic subjectivism. In this regard, our study complemented the patient-reported alopecia with a more objective dermatology assessment.
In conclusion, our pilot project shows promising results for preventing chemotherapy-induced alopecia in the gynecology oncology population treated with dose dense Carboplatin Paclitaxel. A future multi-institutional prospective cohort study is indicated to look at these findings in a larger patient population.
## Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Values are presented as mean ± standard deviation. Group 1: Carboplatin AUC 5-6 and Paclitaxel 175 mg/m 2 every three weeks. Group 2: Carboplatin AUC 5-6 and weekly Paclitaxel 80 mg/m 2 as dose dense.
[fig] Figure 2: Back-view for one of the patients in Group 2 (dose dense Carboplatin AUC 5-6 and weekly Paclitaxel 80 mg/m 2 ) at start of treatment (A), start of cycle 4 (B), and end of cycle 6 (C).CRediT authorship contribution statementCristina Mitric: Conceptualization, Data curation, Formal analysis. Brian How: Conceptualization, Data curation. Emad Matanes: Conceptualization, Data curation. Zainab Amajoud: Conceptualization, Data curation. Hiba Zaaroura: Conceptualization, Data curation. Hai-Hac Nguyen: Conceptualization, Data curation. Angela Tatar: Conceptualization, Data curation. Shannon Salvador: Conceptualization, Methodology, Project administration, Resources. Walter H. Gotlieb: Conceptualization, Methodology, Project administration, Resources. Susie Lau: Conceptualization, Methodology, Project administration, Resources. [/fig]
[table] Table 1: Patient demographics divided into group 1 (Carboplatin AUC 5-6 and Paclitaxel 175 mg/m 2 every three weeks) and group 2 (dose dense with Carboplatin AUC 5-6 and weekly Paclitaxel 80 mg/m 2 ). [/table]
[table] Table 2: European Organization for Research and Treatment of Cancer Quality of Life Questionnaire version 3.0 (EORTC QLQ-C30) and the Body Image Scale (BIS) questionnaire, raw scores. For the EORTC QLQ-C30, a higher score indicates a better quality of life for the global health status and functional scales. A lower score is preferred for the symptom scale of the EORTC QLQ-C30 and for the BIS questionnaire. [/table]
|
Higher blood pressure in elderly hypertensive females, with increased arterial stiffness and blood pressure in females with the Fibrillin-1 2/3 genotype
Background: Elderly patients have a relatively high cardiovascular risk due to increased arterial stiffness, elevated blood pressure and decreased amounts of elastin in the arteries. The composition of the media layer in the arterial wall, comprising elastin, collagen, smooth muscle cells, proteoglycans, fibronectin and fibrillin-1, influences its mechanical properties. Mutations in the fibrillin-1 gene leads to increased aortic stiffness, elevated pulse pressure and aortic root dilatation. This study investigates whether there is a sex difference among hypertensive elderly patients regarding blood pressure, arterial stiffness and fibrillin-1 genotypes. Methods: A total of 315 hypertensive subjects (systolic blood pressure > 140 mmHg) were included in this study (155 men and 160 women aged 71-88 years). Aortic pulse wave velocity and augmentation index were determined using SphygmoCor, and brachial blood pressure was measured using an oscillometric technique. Fibrillin-1 was genotyped by polymerase chain reaction and with a capillary electrophoresis system. Results: Females showed a significantly higher peripheral mean arterial pressure (females; 107.20 mmHg, males 101.6 mmHg, p = 0.008), central mean arterial pressure (females; 107.2 mmHg, males 101.6 mmHg p = 0.008), central systolic blood pressure (females; 148.1 mmHg, males 139.2 mmHg, p < 0.001) and central pulse pressure (females; 68.9 mmHg, males 61.6 mmHg, p = 0.035) than males. Females with the Fibrillin-1 2/3 genotype showed a significantly higher augmentation index (FBN1 2/3; 39.9%, FBN1 2/2 35.0%, FBN1 2/4 35.8, p = 0.029) and systolic blood pressure (FBN1 2/3; 174.6 mmHg, FBN1 2/2168.9 mmHg, FBN1 2/4169.9 mmHg, p = 0.025) than females with the 2/2 and 2/4 genotypes. Conclusion: The findings of this study may indicate that hypertensive elderly females, especially elderly females with Fibrillin-1 2/3, have increased systolic blood pressure and arterial stiffness.
# Background
It is still unclear whether arterial wall stiffness is a cause or a consequence of high blood pressure but the two conditions are close associated to each other. Increased resistance in the small arteries enhance mean blood pressure which increase arterial wall stiffness and pulse wave velocity in the large elastic arteries. This raises central blood pressure further which might lead to organ injuries and damage to the small resistance vessels. Hypertension is a common disease among elderly individuals and is an increasing public health concern. Thus, it is important to identify new potential biomarkers to detect preclinical disease and prevent hypertension. Arterial wall stiffness is an independent predictor of cardiovascular morbidity and mortality, increasing with age and is observed among elderly. An increased arterial stiffness and pulse wave velocity (PWV) cause an early return of the reflected pulse wave in late systole, which increases the central pulse pressure, myocardial oxygen demand and left ventricular workload. Some clinical manifestations of decreased arterial distensibility are hypertension and elevated pulse pressure (PP).
Similar to arterial stiffness, systolic blood pressure (SBP) increases with age. Elderly patients have a higher cardiovascular risk, through increased arterial wall stiffness, elevated blood pressure and decreased levels of elastin in their large elastic arteries. Males have a higher SBP than females throughout most of their lives; however, after 65 years of age, the difference between sexes seems to disappear, and after the age of 70, females may have a higher prevalence of hypertension. Thus, it is of interest to study the differences in the mechanical properties of arteries between the sexes among elderly patients with high SBP.
Mutations in the fibrillin-1 (FBN1) gene is associated with increased aortic stiffness, elevated PP and aortic root dilatation. Several studies have found an association between the FBN1 2/3 genotype and arterial stiffness in males.andreported a relationship between the 2/3 genotype and an increased PP, and in a later study an association between the 2/3 genotype and an increased heart rate (HR), blood pressure and abdominal aortic stiffness in apparently healthy first-degree male relatives of patients with abdominal aortic aneurysms (AAAs) was reported. In a study of both sexes, no associations between the FBN1 2/3 genotype and increased aortic PWV were observed. The aim of this study was to investigate whether there is a sex difference regarding blood pressure, arterial stiffness and the FBN1 genotypes in an elderly hypertensive population.
# Methods
## Study population
The study population was recruited from an ongoing longitudinal study of elderly community-living people from a rural municipality of southeast Sweden. Inhabitants aged 65-82 years were invited to join the original study between 1998 and 2001. A total of 1130 individuals were invited, and 876 of these individuals agreed to participate in the study. In a follow-up study, from 2003 to 2005, a cross-sectional study with the same population was performed. Before the follow-up study started, 123 study subjects had died. Of the remaining 753 subjects, 452 subjects agreed to participate in the current study. Twenty-three subjects were excluded because of irregular HR with variable PP in the arteries, hepatitis infection or difficulties obtaining blood samples. Among the remaining 429 subjects, the abdominal aorta was examined using ultrasound. After the ultrasound examination, an additional 23 subjects were excluded due to low-quality ultrasonic measurements. In this study, we investigated subjects with SBP > 140 mmHg, and the final study population was 315 subjects (155 men and 160 women) within the age group of 71-88 years.
The subjects were instructed not to use tobacco or consume coffee or tea for at least 4 h prior to their examinations. In addition, the participants had a physical examination and were asked about cardiovascular risk factors, cardiovascular diseases and medications. Subjects with previously diagnosed hypertension or with an SBP ≥ 140 mmHg were defined as hypertensive. Diabetes mellitus was defined as a previous diagnosis of diabetes mellitus or with ≥7 mmol l − 1 fasting plasma glucose. Ischemic heart disease was defined as a previous diagnosis of ischemic heart disease confirmed by exercise ECG or coronary angiography or electrocardiographically (ECG) verified myocardial infarction. Subjects who reported that they smoked were classified as smokers. Height and weight were measured and used for calculation of body mass index (BMI). All subjects in the study gave their written informed consent. The study was approved by the Regional Ethical Review board in Linköping, Sweden, and was conducted in accordance with the principles stated in the Declaration of Helsinki.
## Measurements
All measurements were performed under standardized conditions. Brachial blood pressure was measured after at least 15 min of rest when the subject was in the supine position using an oscillometric technique (Dinamap model PRO 200 Monitor, Critikon, Tampa, FL, USA). Systolic, diastolic and mean arterial pressures were measured.
The ankle brachial pressure index (ABPI) was calculated using the SBP measured from the dorsalis pedis and the posterior tibial arteries recorded by a hand-held 8 MHz Doppler probe and a 12-cm standard cuff divided by the brachial systolic pressures measured in the right and left arm.
Radial artery pulse waves were obtained noninvasively using the SphygmoCor system (version 7.0, Model MM3, AtCor Medical, Sydney, Australia). The Sphygmo-Cor system was equipped with a Millar pressure tonometer. For PWV, the central pressure waveform was derived using a generalized transfer function, calculated from an 11-s recording of the radial artery pressure. For calibration of the pressure wave, brachial systolic and diastolic blood pressure were used. Blood pressure was measured before the pulse wave recordings. Registrations of pulse waves were repeated at least three times, and average data from the three most successful registrations were used for analysis. Only good-quality recordings were used for pulse wave analysis (PWA). The augmentation index (AIx) and pressure augmentation (AG) were calculated from the central pressure waveform. AIx is the pressure augmentation expressed as the percentage of the central pulse pressure (cPP): AIx = ( AG/PP) × 100.
## Blood samples
Blood samples were taken after overnight fasting and collected in prechilled plastic Vacutainer tubes (Terumo EDTA K-3). Plasma was prepared by centrifugation at 3000 g for 10 min at 4°C. Blood and plasma were stored at − 70°C pending analysis. Plasma levels of fasting glucose, low-density lipoprotein (LDL) and high-density lipoprotein (HDL) were measured.
## Fbn1 genotyping
FBN1 genotyping was performed in 2009, and the investigators performing these assays were blinded to the results of the vascular testing. DNA was prepared for polymerase chain reaction (PCR) by typing the variable tandem nucleotide repeat (VNTR) (TAAAA) n in intron 28 of the FBN1 gene on chromosome 15 using the forward primer 5′ 6FAM -CAG AGT ACA TAG AGT GTT TTA GGG AGA − 3′ and the reverse primer 5′-GTT TCT TCC TGG CTA CCA TCC AAC TCC c-3′. PCR was performed in a volume of 12 μl with predenaturation at 95°C for 9 min; 35 cycles of denaturation for 30 s, annealing at 65°C for 30 s (35 cycles), and extension at 72°C for 30 s (35 cycles); and a final extension at 72°C. A 1-μl portion of the PCR product was diluted with 9 μl of highly deionized formamide (GENESCAN™ 500ROX™ Size Standard) for electrokinetic injection on the capillary electrophoresis system. The DNA fragments were labeled with the ROX™ fluorophore. The labeling results in a single peak under denaturing conditions were obtained, and the alleles were then identified by the number of base pairs (bp) comparable to each peak (2/2: 162 bp, 162 bp; 2/3: 157 bp, 162 bp; and 2/4: 152 bp, 162 bp).
## Statistics
All data are presented as the means±SDs. All continuous variables were checked for normality distribution, a logtransformation was used in continuous variables with a non-normality distribution. Student's t-test and Mann-Whitney U-test were used to calculate differences between sexes. One-way ANOVA with post hoc testing was used to compare the differences between the genotypes. Variance analysis was used to adjust the dependent variables (peripheral mean arterial pressure (pMAP), central mean arterial pressure (cMAP), central SBP (cSBP) and cPP for age, body surface area (BSA) and HR). p-values < 0.05 were considered statistically significant. The statistical analyses were performed with IBM SPSS 25.0.
# Results
## Demographic data
A total of 315 hypertensive subjects (155 males/160 females; mean age, 78 ± 3.4 years) were analyzed, and demographic data according to sex are presented in. There were no significant differences between sexes regarding age, BMI, CRP, glucose, diabetes, hyperlipidemia or angina pectoris. Male subjects had a higher frequency of smokers (p = 0.005) and higher creatinine (p < 0.001) than female subjects, whereas females had higher HDL (p < 0.001) and LDL (p = 0.048) than males. Males were more affected by myocardial infarction (p = 0.021) than females. However, regarding heredity for cardiovascular disease, females showed increased family history of CV (p = 0.026) compared to males. Furthermore, there were no significant differences between males and females regarding treatments with statins, beta receptor blockers and ACE inhibitors. However, females were more commonly treated with diuretics than males (p = 0.045).
## Blood pressure and arterial stiffness
Data on pressure and arterial stiffness divided by sex are presented in. When the SphygmoCor technique was used, the data showed that females had significantly higher pMAP (p = 0.008) and cMAP (p = 0.008) than males. These sex differences were also observed for cSBP (p < 0.001) and cPP (p = 0.035). After adjustments for age, HR and BSA, these differences were strengthened (pMAP p = 0.005; cMAP p = 0.005; cSBP p = 0.001; cPP p < 0.001). Differences regarding AIx (p < 0.001), AG (p < 0.001) and transit time (TR) (p < 0.001) arterial stiffness were observed between sexes.
Using the oscillometric technique, the SBP was higher among female subjects than among male subjects (p = 0.003), and this difference was maintained after adjustments for age, HR and BSA (p = 0.006). Regarding diastolic blood pressure, no significant difference was found between sexes.
## Fbn1
Four alleles and eight combinations of FBN1 genotypes were identified (1/2, 1/4, 2/2, 2/3, 2/4, 3/3, 3/4, and 4/4), and the three most common genotypes (2/2, 2/3 and 2/ 4) accounted for 79.4% of the population (males 75.4%, females 83.1%). Subsequent analysis was restricted to individuals with the three most common FBN1 genotypes 2/2, 2/3 and 2/4 and were divided by sex. Data on pressure and arterial stiffness according to FBN1 genotype and sex are presented in Tables 3 and 4. The frequencies of the genotypes were 2/2: 60.7% (males), 61.7% (females); 2/3: 14.5% (males), 13.5% (females); and 2/4: 34.2% (males), 24.8% (females).
There was a significant difference in HR between the three main genotypes in the male group. The 2/2 genotype showed a significantly higher HR (p = 0.036) than the 2/3 and 2/4 genotypes. After adjustments for age, the statistically significant difference compared to the 2/3 and 2/4 genotypes disappeared.
In the female group, the 2/3 genotype showed a significantly higher AIx (p = 0.029) and SBP (p = 0.025) than the 2/2 and 2/4 genotypes. The difference in AIx between the genotypes was still significant after adjustment for age (p = 0.031). Regarding SBP, the 2/3 genotype showed a slight increase in SBP after adjustment for age; however, no significant difference between the 2/3 genotype and the 2/2 or 2/4 genotype was observed. Furthermore, the 2/3 genotype in females had a significantly lower HR (p = 0.048) than the 2/2 and 2/4 genotypes.
# Discussion
This study shows that elderly hypertensive females have a higher SBP, pMAP, cMAP, AIx, AG and TR than elderly hypertensive males. We have also shown a higher SBP and a higher AIx among hypertensive elderly females with the FBN1 2/3 genotype compared to hypertensive females with the two other common genotypes (2/2 and 2/4). Small but not statistical significantly differences in AIx between the FBN1 genotypes were also seen in male. Increased arterial stiffness results in elevated systolic blood pressure, which is a prognostic indicator of future cardiovascular events. However, the Framingham study showed an increase in the incidence of cardiovascular diseases in females with age, and the cardiovascular incidence seems to be higher after menopause. The lower level of estrogen in postmenopausal women could be the main reason for the increased risk of cardiovascular diseases. Many postmenopausal females are treated with hormone replacement therapy (HRT), and Mueck et al. found that both normotensive and hypertensive postmenopausal women run a very low risk of blood pressure increase during HRT.
Increases in SBP, cSBP, central and peripheral MAP, and cPP were observed in the hypertensive elderly females of our study compared to males, and these findings are in accordance with those of previous reports. We used an oscillometric technique to measure brachial blood pressure, and a noninvasive tonometric system was used to analyze the pulse wave of the radial artery. Furthermore, when blood pressure was adjusted for age, HR and BSA, the differences between the sexes increased.used invasive measurements of blood pressure and observed the same trend in a large cohort, with a higher SBP and AIx (p = 0.048) among elderly females than among elderly males, similar to our findings . Elderly hypertensive females seem to have higher SBP than males; thus, it is important to discuss the differences between sexes regarding the treatment of hypertension. Ong et al. (2008) studied sex differences in blood pressure rate control among Americans with diagnosed hypertension and showed that blood pressure measurement control rates were not significantly lower among females than among males. The guidelines recommend that hypertensive males and females should be treated with the same approach, and evidence shows that females show a benefit from antihypertensive treatment that is similar to that of males. The 2018 ESH/ESC Guidelines recommend the use of antihypertensive drugs by all patients diagnosed with grade 1 hypertension. The use of antihypertension drugs may affect the arterial impedance. In the present study we found a significance difference between sexes in diuretic treatment. However, diuretics seem to have no beneficial effect on pulsatile hemodynamics and a neutral effect on AIx and PWV. Beta-Blockers, ACE-inhibitors and Calcium channel blockers have been seen to improves arterial stiffness, but there was no significance difference between sexes in treatment with these medications.
FBN1 is a glycoprotein that forms FBN1-rich microfibrils and functions as a skeleton for the deposition of tropoelastin, forming elastic fibers in the arterial wall. FBN1 might affect the maintenance of the integrity of the arterial wall of central arteries, and FBN1 is considered to be a candidate gene of arterial stiffness. The different FBN1 genotypes have been studied in several populations, and the frequencies of these genotypes in our present study are in accordance with what others have reported previously among healthy subjects, AAA-patients, patients with coronary artery diseasesand middle-aged subjects. The increased prevalence of the FBN1 2/3 genotype in this population may explain the impact on the structure of the aortic wall, as shown by increased aortic stiffness.
The females in the present study with the FBN1 2/3 genotype had a higher AIx and SBP than those with the other common genotypes, namely, 2/2 and 2/4. These differences were not observed in males. One causative explanation of the observed sex difference could be the fact that hypertensive males with high levels of and/or multiple risk factors for cardiovascular diseases already have died in cardiovascular events due to the early onset of cardiovascular morbidity in males. Arterial stiffness is an independent predictor of cardiovascular morbidity and mortality, and AIx is an indirect measure of arterial stiffness. Several studies have found an association between the FBN1 2/3 genotype and arterial stiffness in males; however, we observed an association between higher AIx and FBN1 2/3 in only elderly hypertensive females and not in elderly hypertensive males. Furthermore, after adjustment for age, the increase in AIx was reduced, although it still reached significance. In studies of both sexes (healthy individuals), no associations between the FBN1 2/3 genotype and arterial stiffness were observed. Compared to that of the previously published work, our population consisted of hypertensive elderly patients, and this could be a possible explanation for the difference in findings. Arterial aging is associated with a decreased amount of elastin in the large elastic arteries and affects arterial stiffness. As arterial stiffness is an independent predictor of cardiovascular morbidity and mortality, some elderly hypertensive patients with high levels of and/or multiple risk factors for cardiovascular diseases may have already died due to cardiovascular events. Thus, it must be considered that the differences between the genotypes may be larger in a younger population.
# Limitations
The interpretation of the findings from the present study has limitations. First, the different number of subjects in each FBN1 genotype groupcould influence the results due to limited numbers of subjects in the FBN1 2/3 genotype group. Second, the PWA was calculated from measurements of the radial artery and was performed according to the manufacturer's recommendations and guidelines regarding measurements of PWA. However, peripheral blood pressure may not accurately reflect actual aortic PWV. Third, our analysis follows a cross-sectional design and no data on outcomes are reported; therefore, no direct cause-and-effect associations can be derived from this study design. A fourth limitation is that no lifestyle items, medical history of stroke and information about diabetic microvascular complications were included. Our participants were old and the duration of diabetes and HbA1c was not included in the study. Finally, it is a limitation that we do not have data about left ventricular ejection fraction and valvular heart disease. More studies are needed to define the possible effect that the FBN1 2/3 genotype may have on cardiovascular morbidity and mortality.
# Conclusion
In conclusion, the findings of our study indicate that hypertensive females in an elderly population have increased SBP and arterial stiffness compared to males. These findings also indicate that elderly females with FBN1 2/3 have increased arterial stiffness compared to hypertensive elderly females with FBN1 2/2 or FBN1 2/ 4. Therefore, these findings may be used to design potential therapeutic targets to reduce hypertension in females with the FBN1 2/3 genotype. Measurement of PWV may be a useful tool in clinical practice, especially in female, to select subjects at high risk of developing cardiovascular diseases. However, further studies are needed to confirm our results. |
Neurocognitive features in subgroups of bipolar disorder
Objective: To examine which subgroups of DSM-IV bipolar disorder (BD) [BD type I (BD-I) or BD type II (BD-II), and subgroups based on history of psychosis, presenting polarity, and age at onset] differentiate best regarding neurocognitive measures.Methods: A total of 199 patients with BD were characterized by clinical and neurocognitive features. The distribution of subgroups in this sample was: BD-I, 64% and BD-II, 36%; 60% had a history of psychosis; 57% had depression as the presenting polarity; 61% had an early onset of BD, 25% had a mid onset, and 14% had a late onset. We used multivariate regression analyses to assess relationships between neurocognitive variables and clinical subgroups.Results: Both BD-I diagnosis and elevated presenting polarity were related to impairments in verbal memory, with elevated presenting polarity explaining more of the variance in this cognitive domain (22.5%). History of psychosis and BD-I diagnosis were both related to impairment in semantic fluency, with history of psychosis explaining more of the variance (11.6%).Conclusion: Poor performance in verbal memory appears to be associated with an elevated presenting polarity, and poor performance in semantic fluency appears to be associated with a lifetime history of psychosis.
Bipolar disorder (BD) is a severe mental disorder with marked heterogeneity in symptomatology, treatment response, clinical course, and outcome [bib_ref] Functioning and disability in bipolar disorder: an extensive review, Sanchez-Moreno [/bib_ref] [bib_ref] Predicting the course and outcome of bipolar disorder: a review, Treuer [/bib_ref] [bib_ref] Diagnostic features, prevalence, and impact of bipolar disorder, Ketter [/bib_ref]. In addition to the presence of severe mood episodes, the disorder is also associated with varying degrees of psychotic symptoms, neurocognitive impairments, and loss of functioning. This heterogeneity within BD has led to considerable efforts to establish more homogeneous subgroups to use in the search for genetic underpinnings, pathogenic factors, and mechanisms behind treatment response. The DSM-IV comprises two separate subgroups of BD that are exclusively based on differences in their severity of mood elevation; here BD type I (BD-I) is characterized by mania, while BD type II (BD-II) is characterized by hypomania.
Recently, an increased focus on possible BD endophenotypes has emerged from genetic studies. This focus has been emphasized in ongoing work on revisions of the current diagnostic systems (i.e., . In this context, proposals for novel subgroups within the BD spectrum have appeared [bib_ref] Genetic research into bipolar disorder: the need for a research framework that..., Schulze [/bib_ref]. These include differentiation between psychotic and non-psychotic BD, and between depressive and elevated polarity of the first (or presenting) episode, in addition to groups based on differences in age at onset. However, no studies have so far assessed to what extent these subgroups overlap within one sample, or their ability to discriminate between groups differing in significant characteristics unrelated to those inherent in the grouping procedure, such as cognition.
A recent large international study suggests lifetime prevalence in the general population of 0.6% for BD-I and 0.4% for BD-II [bib_ref] The nosologic validity of paranoia (simple delusional disorder). A review, Kendler [/bib_ref]. Comparative studies indicate that BD-I is associated with psychotic episodes [bib_ref] Prevalence and correlates of bipolar spectrum disorder in the world mental health..., Merikangas [/bib_ref] and hospitalizationsto a greater extent than BD-II. On the other hand, patients with BD-II are more likely to experience and spend more time in depressive episodes than BD-I patients [bib_ref] Functioning and disability in bipolar disorder: an extensive review, Sanchez-Moreno [/bib_ref]. There are, however, no differences in demographic characteristics, age at onset [bib_ref] Bipolar II illness: course and outcome over a five-year period, Coryell [/bib_ref] , functional outcome [bib_ref] The comparative clinical phenotype and long term longitudinal episode course of bipolar..., Judd [/bib_ref] , or rate of suicide attempts between BD-I and BD-II [bib_ref] Functional impairment in bipolar II disorder: is it as disabling as bipolar..., Rosa [/bib_ref].
Psychotic symptoms are reported in about 70% of patients with BD-I [bib_ref] Suicide attempts in bipolar I and bipolar II disorder: a review and..., Novick [/bib_ref] [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref] and 20% of those with BD-II [bib_ref] Impact of psychotic features on morbidity and course of illness in patients..., Ozyildirim [/bib_ref] , with no gender differences [bib_ref] Suicide attempts in bipolar I and bipolar II disorder: a review and..., Novick [/bib_ref] [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref]. There are indications that psychotic BD constitutes a subgroup with a higher frequency of elevated (manic and hypomanic) mood episodes, more severe mood episodes, more hospitalizations [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref] , and more cognitive impairments [bib_ref] Psychotic versus non-psychotic bipolar II disorder, Mazzarini [/bib_ref] [bib_ref] Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study, Bora [/bib_ref] compared to non-psychotic BD. A previous study from our research group also indicated that psychotic BD predicted neurocognitive dysfunction [bib_ref] Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study, Bora [/bib_ref] to a larger extent than a BD-I diagnosis, although others do not find this association [bib_ref] Neurocognitive profiles in bipolar I and bipolar II disorder: differences in pattern..., Simonsen [/bib_ref]. Patients with non-psychotic BD appear, on the other hand, to have more first-degree relatives with BD [bib_ref] Suicide attempts in bipolar I and bipolar II disorder: a review and..., Novick [/bib_ref] [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref] , more depressive episodes, and a better response to treatment with lithium [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref].
For about one-half [bib_ref] Bipolar I patients with and without a history of psychotic symptoms: do..., Selva [/bib_ref] to two-thirds [bib_ref] Polarity of the first episode, clinical characteristics, and course of manic depressive..., Perugi [/bib_ref] of cases, the first presenting episode is depression. Studies show either equal sex distribution [bib_ref] Polarity of the first episode, clinical characteristics, and course of manic depressive..., Perugi [/bib_ref] [bib_ref] Clinical correlates of first-episode polarity in bipolar disorder, Daban [/bib_ref] or a higher frequency of female patients with depressive onset [bib_ref] Bipolar I patients with and without a history of psychotic symptoms: do..., Selva [/bib_ref]. At least in BD-I, the polarity of presenting episode has been shown to be a feature running in families [bib_ref] The soft bipolar spectrum redefined: focus on the cyclothymic, anxious-sensitive, impulsedyscontrol, and..., Perugi [/bib_ref] , and there are indications that a depressive presenting polarity is associated with earlier age at onset [bib_ref] Familiality of polarity at illness onset in bipolar affective disorder, Kassem [/bib_ref] [bib_ref] Polarity at illness onset in bipolar I disorder and clinical course of..., Forty [/bib_ref] , more frequent episodes, and a predominant polarity of depression throughout the course of the disorder [bib_ref] Bipolar I patients with and without a history of psychotic symptoms: do..., Selva [/bib_ref] [bib_ref] Familiality of polarity at illness onset in bipolar affective disorder, Kassem [/bib_ref] [bib_ref] Polarity at illness onset in bipolar I disorder and clinical course of..., Forty [/bib_ref] [bib_ref] Revisiting depressive-prone bipolar disorder: polarity of initial mood episode and disease course..., Perlis [/bib_ref]. Others have found similar patterns, but with a later age at onset in depressive presenting polarity [bib_ref] Polarity of the first episode, clinical characteristics, and course of manic depressive..., Perugi [/bib_ref]. Patients with depressive presenting episodes also appear to have longer treatment delays [bib_ref] Clinical and evolutionary characteristics of bipolar disorder according to the polarity of..., Bram [/bib_ref] [bib_ref] Polarity of the first episode and time to diagnosis of bipolar I..., Cha [/bib_ref] and more suicide attempts [bib_ref] Clinical and evolutionary characteristics of bipolar disorder according to the polarity of..., Bram [/bib_ref] [bib_ref] Polarity of the first episode and time to diagnosis of bipolar I..., Cha [/bib_ref] [bib_ref] Does first episode polarity predict risk for suicide attempt in bipolar disorder?, Chaudhury [/bib_ref] than patients with elevated or mixed onsets. Polarity of presenting episode could inform treatment, as it may anticipate predominant polarity and thus the most effective medical treatment [bib_ref] Initial depressive episodes affect the risk of suicide attempts in Korean patients..., Ryu [/bib_ref].
The age range for the onset of BD is very wide [bib_ref] New ways to classify bipolar disorders: going from categorical groups to symptom..., Henry [/bib_ref] , with no gender differences [bib_ref] Age at onset of bipolar disorder in a Norwegian catchment area sample, Larsson [/bib_ref]. Previous research has focused on early versus late onset, but recent large multisite studies have identified three potential age at onset-based subgroups with different clinical presentation, across different cultural settings and birth cohorts [bib_ref] Is sex important? Gender differences in bipolar disorder, Diflorio [/bib_ref] [bib_ref] Age at onset in bipolar I affective disorder: further evidence for three..., Bellivier [/bib_ref] [bib_ref] Age-atonset in bipolar-I disorder: mixture analysis of 1369 cases identifies three distinct..., Hamshere [/bib_ref] [bib_ref] Age at onset in bipolar affective disorders: a review, Leboyer [/bib_ref] [bib_ref] Age at onset in Sardinian bipolar I patients: evidence for three subgroups, Manchia [/bib_ref] : i.e., early onset (mean age at onset % 17 years), intermediate onset (mean age at onset % 26 years), and late onset (mean age at onset % 42 years). The characteristics of the disorder may vary with the age at onset [bib_ref] New ways to classify bipolar disorders: going from categorical groups to symptom..., Henry [/bib_ref] , and those with an early onset appear as a separate subgroup with specific clinical manifestations including higher recurrence rates of mood episodes, more elevated episodes at least in BD-I (36), more often depressive onsets [bib_ref] Age at onset in Sardinian bipolar I patients: evidence for three subgroups, Manchia [/bib_ref] , more suicide attempts [bib_ref] Age at onset in bipolar I affective disorder: further evidence for three..., Bellivier [/bib_ref] [bib_ref] Age at onset in Sardinian bipolar I patients: evidence for three subgroups, Manchia [/bib_ref] , higher risk for comorbid borderline personality disorder [bib_ref] Classifying mood disorders by age-at-onset instead of polarity, Benazzi [/bib_ref] , higher rates of psychotic symptoms [bib_ref] Age at onset of bipolar disorder and risk for comorbid borderline personality..., Goldberg [/bib_ref] [bib_ref] Phenomenology and outcome of subjects with early-and adult-onset psychotic mania, Carlson [/bib_ref] , more frequent neurocognitive impairment (40), more BD-I than BD-II [bib_ref] Age at onset in Sardinian bipolar I patients: evidence for three subgroups, Manchia [/bib_ref] , and more often a family history of BD compared to patients with later onsets.
Thus, there is considerable empirical evidence of subgroups in BD that are associated with differences in clinical course and outcome. There has, however, been limited attention paid to the possibility that the different subgroups describe overlapping phenomena, as indicated by several characteristics common to the suggested groups. Even if we focused here on different aspects or subgroups of a specific disorder, some of the same validating principles should apply here, as for disease entities. Suggested validation criteria for psychiatric illness can be divided into three major categories [bib_ref] Bipolar disorder. II: personality and age of onset, Engstrom [/bib_ref] : antecedent validators (family history, demographic, and precipitating factors); concurrent validators (psychological factors derived from, for example, symptom interviews); and predictive validators (relapse, treatment response, and other course descriptions). As the definitions of the proposed subgroups encompass either antecedent and predictive characteristics (age at onset, type, and order of episodes) or antecedent and concurrent clinical syndromes (for BD-I/BD-II and history of psychosis), differentiation based on these characteristics may increase the risk for tautological conclusions. A step forward could be to show subgroup differences in concurrent characteristics that are not directly associated with criteria for subgroup formation. One candidate here is neurocognitive functioning.
Cognitive impairments are present already in the early course of BD [bib_ref] Relationship between cognitive functioning and 6-month clinical and functional outcome in patients..., Torres [/bib_ref] and are an important determinant of functional outcome [bib_ref] Neurocognitive impairment in bipolar disorder patients: functional implications, Wingo [/bib_ref]. Metaanalyses provide evidence of trait-like neuropsychological deficits in BD involving impairments in attention, processing speed, memory, and executive function [bib_ref] Neuropsychological functioning in euthymic bipolar disorder: a meta-analysis, Torres [/bib_ref]. Comparative studies suggest that BD-I is characterized by reduced cognitive performance compared to BD-II on executive function [bib_ref] Are BPI and BPII suicide attempters distinct neuropsychologically?, Harkavy-Friedman [/bib_ref] [bib_ref] Neuropsychological functions in patients with bipolar I and bipolar II disorder, Hsiao [/bib_ref] [bib_ref] Cognitive impairment in bipolar II disorder, Torrent [/bib_ref] , verbal [bib_ref] Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study, Bora [/bib_ref] [bib_ref] Neuropsychological functions in patients with bipolar I and bipolar II disorder, Hsiao [/bib_ref] [bib_ref] Cognitive impairment in bipolar II disorder, Torrent [/bib_ref] and working memory [bib_ref] Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study, Bora [/bib_ref] , and processing speed [bib_ref] Neuropsychological functions in patients with bipolar I and bipolar II disorder, Hsiao [/bib_ref]. A recent meta-analytic review also concluded that although BD-II patients are less impaired than BD-I patients on memory and semantic fluency, the overall cognitive impairment in BD-II appears as severe as in BD-I (52). Psychotic BD has been shown to be associated with more impairments than non-psychotic BD, in relation to executive function in general [bib_ref] Psychotic versus non-psychotic bipolar II disorder, Mazzarini [/bib_ref] [bib_ref] The neurocognitive signature of psychotic bipolar disorder, Glahn [/bib_ref] [bib_ref] Neurocognitive impairment in bipolar patients with and without history of psychosis, Martinez-Aran [/bib_ref] [bib_ref] Neurocognitive dysfunction in bipolar and schizophrenia spectrum disorders depends on history of..., Simonsen [/bib_ref] , and cognitive flexibility in particular [bib_ref] The effect of previous psychotic mood episodes on cognitive impairment in euthymic..., Bora [/bib_ref] [bib_ref] Neuropsychological status of bipolar I disorder: impact of psychosis, Savitz [/bib_ref] , as well as on measures of verbal memory [bib_ref] Psychotic versus non-psychotic bipolar II disorder, Mazzarini [/bib_ref] [bib_ref] Neurocognitive impairment in bipolar patients with and without history of psychosis, Martinez-Aran [/bib_ref] [bib_ref] Neurocognitive dysfunction in bipolar and schizophrenia spectrum disorders depends on history of..., Simonsen [/bib_ref] , working memory [bib_ref] Psychotic versus non-psychotic bipolar II disorder, Mazzarini [/bib_ref] [bib_ref] The neurocognitive signature of psychotic bipolar disorder, Glahn [/bib_ref] [bib_ref] Neurocognitive dysfunction in bipolar and schizophrenia spectrum disorders depends on history of..., Simonsen [/bib_ref] [bib_ref] Neuropsychological status of bipolar I disorder: impact of psychosis, Savitz [/bib_ref] , and processing speed [bib_ref] Psychotic versus non-psychotic bipolar II disorder, Mazzarini [/bib_ref]. For other suggested subgroups, the data on cognitive features are rather limited. Early onset may be associated with more severe impairments in verbal memory and processing speed [bib_ref] Adolescent versus adult onset of mania, Mcglashan [/bib_ref]. Since neurocognitive impairments are not defining features of clinical subgroups, they can serve as important concurrent validators.
The aim of the current study was to examine to what extent the different suggested ways of subgrouping BD influence cognitive test results in areas of neurocognition previously implicated in BD.
# Materials and methods
## Participants
All participants were consecutively recruited to the ongoing Thematically Organized Psychosis (TOP) Study from outpatient and inpatient units of the three major hospitals in Oslo, Norway, between the years 2003 and 2009. The treating staff asked patients if they were interested in participating in a study of BD, and if so, they were referred to the study. Inclusion criteria for this particular study were age between 17 and 65 years and having a DSM-IV diagnosis of BD-I or BD-II [Total N = 199: BD-I (n = 128), BD-II (n = 71)]. The participants were required to have a Scandinavian language as their first language or have received their compulsory schooling in Scandinavia to assure valid neurocognitive test performance. General exclusion criteria were hospitalization for a head injury, neurological disorder, unstable or uncontrolled medical condition that interferes with brain function, and/or an IQ below 70. The Regional Committee for Medical Research Ethics and the Norwegian Data Inspectorate approved the study, and participants' written, informed consent according to the Declaration of Helsinki was obtained.
The sample consisted of 117 female patients (59%) and 82 male patients (41%); 71 (36%) were employed; 69 (35%) were married or cohabitating. A total of 56 patients (28%) had one or more first-degree relatives with BD (n = 43, 22%), schizophrenia (n = 9, 4%), or both schizophrenia and BD (n = 4, 2%). Lifetime suicide attempt was present in 55 patients (28%). None of these variables differed significantly within any of the four subgroups, except that patients with a depressive presenting polarity were more often single than those with elevated presenting polarity (73% versus 56%, respectively, p = 0.017). Mean age for the whole sample was [mean AE standard deviation (SD)] 37 AE 12 years and the median for duration of treatment was one year. There was a significant age difference among the three age at onset groups (mean = 32, 40, and 49 years, respectively, p < 0.001) and between the depressive and elevated presenting polarity groups (mean = 34 versus 39 years, respectively, p = 0.006).
The distribution of the different subgroups was: (i) BD-I (n = 128, 64%) versus BD-II (n = 78, 36%); (ii) psychotic BD (n = 120, 60%) versus non-psychotic BD (n = 78, 40%); (iii) depressive presenting polarity (n = 114, 59%) versus elevated presenting polarity (n = 80, 41%); and (iv) early onset (n = 120, 61%) versus mid onset (n = 49, 25%) and late onset (n = 28, 14%).
## Clinical assessments
Patients were clinically characterized based on a personal interview by trained assessment staff, either medical doctors or clinical psychologists, who had completed the TOP Study's assessment training and reliability program. A good inter-rater reliability for diagnoses was achieved with an overall kappa score of 0.77 (95% confidence interval: 0.60-0.94) [bib_ref] Differences in prevalence and patterns of substance use in schizophrenia and bipolar..., Ringen [/bib_ref]. Diagnosis was based on the Structured Clinical Interview for DSM-IV Axis I disorders (SCID-I) (59) and information from medical charts. History of psychosis, polarity of presenting episode, and age at onset were determined from the same clinical interview, particularly the SCID information on previous psychotic and mood episodes, and from medical charts. A history of psychosis was defined as having one or more lifetime psychotic episodes. We defined polarity of first presenting episode as the polarity of the first SCID-verified mood episode. Only four patients had a mixed episode as first presenting episode and, due to the low number, they were grouped together with the mania/hypomania as first-episode group. Age at onset was defined as the age of the first SCID-verified mood episode. Age was collapsed into three groups based on results from previous admixture analysis in large samples finding relatively stable age at onset groups in different cultures and birth cohorts: (i) early onset (first episode at <22 years), (ii) mid onset (first episode at between 23 and 34 years), and (iv) late onset (first episode at >35 years) [bib_ref] Is sex important? Gender differences in bipolar disorder, Diflorio [/bib_ref] [bib_ref] Age at onset in bipolar I affective disorder: further evidence for three..., Bellivier [/bib_ref] [bib_ref] Age-atonset in bipolar-I disorder: mixture analysis of 1369 cases identifies three distinct..., Hamshere [/bib_ref] [bib_ref] Age at onset in bipolar affective disorders: a review, Leboyer [/bib_ref]. Medication status was based on information from interview and medical charts. Current use of mood-stabilizing medication was reported in 119 patients (60%), antidepressants in 80 patients (40%), and antipsychotic medication in 97 (49%). Treatment onset was defined as the first contact with a specialist, regardless of episode polarity. Family history was based on a semi-structured interview asking patients about the presence of BD or schizophrenia in first-degree relatives (parents and siblings). The patient answered whether the diagnosis was probable or sure (as diagnosed by a doctor). We included both. Patients who were adopted or did not know the identity of their father were excluded from this analysis.
The Positive and Negative Symptom Scale (PANSS) was used to measure current psychotic symptoms. The psychosis cut-off was at a level ! 4 on items p1, p3, p5, p6, and/or g9 [bib_ref] The positive and negative syndrome scale (PANSS) for schizophrenia, Kay [/bib_ref]. There were no differences in rates of current psychosis between any of the subgroups, apart from the history of psychosis subgroup where 22 patients (19%) had current psychotic symptoms. Current depressive symptoms were measured by the Inventory of Depressive Symptoms-Clinician rated (IDS-C) [bib_ref] The Inventory of Depressive Symptomatology (IDS): psychometric properties, Rush [/bib_ref]. Here, 91 patients (48%) had no depressive symptoms (IDS 13), 41 (22%) had possible/ mild depression (IDS score 14-21), 32 (17%) had moderate depression (IDS score 22-30), 16 (8%) had severe depression (IDS score 31-38), and 10 (5%) had very severe depression (IDS score ! 39) (62). IDS information was missing for nine patients. Current manic symptoms were rated using the Young Mania Rating Scale (YMRS) [bib_ref] A rating scale for mania: reliability, validity and sensitivity, Young [/bib_ref]. Here, 166 patients (84%) had no symptoms of mania (YMRS score 7), 29 (15%) had possible/mild mania (YMRS score [bib_ref] Bipolar II illness: course and outcome over a five-year period, Coryell [/bib_ref] [bib_ref] The comparative clinical phenotype and long term longitudinal episode course of bipolar..., Judd [/bib_ref] [bib_ref] Functional impairment in bipolar II disorder: is it as disabling as bipolar..., Rosa [/bib_ref] [bib_ref] Suicide attempts in bipolar I and bipolar II disorder: a review and..., Novick [/bib_ref] [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref] [bib_ref] Impact of psychotic features on morbidity and course of illness in patients..., Ozyildirim [/bib_ref] [bib_ref] Psychotic versus non-psychotic bipolar II disorder, Mazzarini [/bib_ref] [bib_ref] Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study, Bora [/bib_ref] [bib_ref] Neurocognitive profiles in bipolar I and bipolar II disorder: differences in pattern..., Simonsen [/bib_ref] [bib_ref] Bipolar I patients with and without a history of psychotic symptoms: do..., Selva [/bib_ref] [bib_ref] Polarity of the first episode, clinical characteristics, and course of manic depressive..., Perugi [/bib_ref] [bib_ref] Clinical correlates of first-episode polarity in bipolar disorder, Daban [/bib_ref] , and two (1%) had moderate mania (YMRS score 21-30) (62). YMRS information was missing for two patients. Eighty-six patients (36%) were euthymic (IDS score 13 and YMRS score 7).
Premorbid adjustment was measured by the Premorbid Adjustment Scale (PAS) [bib_ref] Measurement of premorbid adjustment in chronic schizophrenia, Cannon-Spoor [/bib_ref] , and then subdivided into the domains of social and academic adjustment using indices of childhood level and subsequent change, up to the last premorbid period [bib_ref] First-episode psychosis: diagnostic stability over one and two years, Haahr [/bib_ref]. A higher PAS score indicates a lower functioning. Functional and symptomatic levels were assessed with the Global Assessment of Functioning (GAF) Scale, split version [bib_ref] Generalizability studies of the Global Assessment of Functioning-Split version, Pedersen [/bib_ref].
## Neurocognitive assessment
Psychologists trained in standardized neuropsychological testing carried out neurocognitive assessment. A three-hour comprehensive test battery was administered in a fixed order with two breaks with refreshments. Premorbid IQ was assessed with a Norwegian research version of the National Adult Reading Test (NART) [bib_ref] Estimating premorbid IQ (In Norwegian with English abstract), Sundet [/bib_ref]. There were no differences in premorbid IQ within any of the four subgroups.
Included in this part of the study were neurocognitive tests, which measure cognitive functions sensitive to BD [bib_ref] Neurocognitive markers of psychosis in bipolar disorder: a meta-analytic study, Bora [/bib_ref] (71), including both semantic and phonemic fluency, with additional measures of repetition and set loss errors. Verbal interference control was measured through the inhibition trial, and interference set shifting through the inhibition switching trial, of the Color-Word Interference Test (D-KEFS), with additional information about the number of inhibition and inhibitionswitching errors. We used raw scores on all tests.
## Statistical analyses
All analyses were performed using The Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL, USA; version 18.0). Bivariate analyses investigating differences between groups (Tables 1-4) used v² tables for categorical data, Mann-Whitney U-tests and Kruskal-Wallis H-tests for non-normally distributed continuous data, and t-tests and ANOVAs for normally distributed data. The level of statistical significance with Bonferroni correction for multiple testing was set to p 0.017. To be able to adjust for potential mediators (variables correlated with both subgrouping and outcome variables), we performed bivariate correlation analyses between demographical and clinical variables, measured through Pearson's correlations (r). Variables explored were sex, age, education, duration of illness, number of episodes, number of depressive episodes, number of elevated episodes, and level of current symptomatology, such as level of depressive and manic symptoms, and presence of psychotic symptoms.
To explore the effect of potential confounders for the association between group membership and neurocognition, we first performed bivariate correlation analyses between group membership, and neurocognitive, demographic, and clinical variables [Pearson's correlations (r)]. Variables explored were sex, age, education, duration of illness, number of episodes, and level of current symptomatology, such as level of depressive and manic symptoms, as well as the presence of psychotic symptoms. We then conducted a series of hierarchical multiple linear regression analyses with neurocognitive variables that showed withingroup differences in at least two of the suggested four subgroups as dependents (i.e., verbal memory, verbal learning, and semantic fluency). We used a block-wise forced entry procedure, and in the first block entered variables with significant association with the dependent in bivariate correlations (i.e., age, sex, duration of illness for verbal memory and education, and age for verbal fluency). In the second block we added affective, psychotic symptoms as well as duration of illness and number of episodes, as these theoretically could affect neurocognitive functioning. The third block contained information on subgroup membership for groups that showed neurocognitive differences for the dependents (i.e., BD-I versus BD-II, history of psychosis, presenting episode, and finally age at onset as a continuous variable, respectively).
# Results
## Group differences in patient characteristics
Diagnostic subgroup (BD-I versus BD-II). Patients with a BD-I diagnosis had lower PAS childhood scores and fewer mood episodes compared to patients with a BD-II diagnosis. In addition, BD-I patients were more often euthymic and had lower GAF scores than BD-II patients. As a group, the BD-II patients also had more depressive symptoms. Patients with BD-I performed significantly worse than patients with BD-II on verbal learning (p 0.001), verbal memory (p 0.001), and semantic fluency (p = 0.010) [fig_ref] Table 1: Clinical and neurocognitive characteristics of patients with bipolar I [/fig_ref]. A larger proportion of the BD-I group used antipsychotic medication (v 2 = 14.92, p 0.001) than the BD-II group, who to a larger extent used antidepressants (v 2 = 12.53, p 0.001).
Psychotic symptoms. Patients with psychotic BD had a shorter duration of illness and had experienced fewer elevated mood episodes than non-psychotic BD patients. Non-psychotic BD patients had, in turn, more depressive symptoms, but had higher GAF scores. They also displayed a trend toward more first-degree relatives with BD (v 2 = 3.71, p = 0.069). Patients with psychotic BD performed significantly worse than those with nonpsychotic BD on verbal memory (p = 0.017) and semantic fluency (p = 0.011) [fig_ref] Table 2: Clinical and neurocognitive characteristics of patients with bipolar disorder with and without... [/fig_ref]. A larger proportion of the psychotic BD group used anti-psychotic medication (v 2 = 29.95, p 0.001) compared to the non-psychotic group, who to a larger extent used antidepressants (v 2 = 11.91, p = 0.001).
Polarity of presenting episode. Patients with a depressive presenting polarity were younger both at onset of disorder and at study entrance than those with an elevated presenting polarity. The group with a depressive presenting polarity also had experienced more depressive mood episodes. However, they performed significantly better than those with an elevated presenting polarity on verbal learning (p 0.001) and verbal memory (p 0.001) and had fever intrusions on the CVLT (p = 0.017) [fig_ref] Table 3: Clinical and neurocognitive characteristics of patients with bipolar disorder with depressive and... [/fig_ref].
Age at onset. The three age at onset groups also differed in current age. The earlier-onset groups had poorer PAS social and school scores, a longer duration of illness, a higher number of both depressive and elevated episodes, and more current depressive symptomatology. After controlling for multiple testing on neurocognitive measures, the three age at onset groups did not differ statistically significantly from each other [fig_ref] Table 4: Clinical and neurocognitive characteristics of patients with bipolar disorder with early, mid,... [/fig_ref].
## Neurocognitive functioning across subgroups
Three neurocognitive measures differed statistically significantly across two or more subgroups, also after correcting for multiple testing: verbal learning, verbal memory, and semantic fluency. Since verbal learning and verbal memory were highly inter-correlated (r = 0.74, p 0.01) and analyses gave similar results, we only report here the results for verbal memory. To investigate the independent explanatory power of the different ways to subgroup, we performed two different multivariate analyses, one with verbal memory and one with semantic fluency as dependent variables (for details of procedure, see 'Statistical analyses' section above).
Possible confounders for the association between verbal memory and subgroups were age, gender, and duration of illness, entered in the first block. Age (p = 0.004) and sex (p = 0.002) significantly contributed to the model. In the second block, neither affective symptoms nor number of episodes affected verbal memory, while current psychotic symptoms (p = 0.003) did. Having a BD-I diagnosis, history of psychosis, and an elevated presenting episode were associated with poorer performance on verbal memory in the bivariate analyses, but all could not be entered in the same model due to collinearity problems. Regarding the effect of subgroups, the best model was the one containing elevated presenting episode followed by BD-I [fig_ref] Table 5: Hierarchical regression model for verbal memory Significant results [/fig_ref].
For the analysis of sematic fluency, age and level of education were possible confounders entered in the first block. Age (p = 0.005) and education (p = 0.010) significantly contributed to the model. In the second block, neither affective nor psychotic symptoms, nor duration of illness, nor number of episodes contributed to the model. Having a BD-I diagnosis and history of psychosis contributed to a poorer verbal fluency. Again, the model did not adequately fit both. The best model contained history of psychosis [fig_ref] Table 6: Hierarchical regression model for semantic fluency Significant results [/fig_ref].
# Discussion
The main finding is that three of the suggested subgroups (BD-I versus BD-II, history of psychosis, and presenting polarity) differed in regard to their association with aspects of neurocognitive functioning; in particular, verbal memory and semantic fluency. It has been suggested that verbal memory impairment is a BD endophenotype, as it seems to be a trait-related deficit (72) that is also present in relatives of patients with BD [bib_ref] Memory functioning in familial bipolar I disorder patients and their relatives, Quraishi [/bib_ref]. It is of particular interest to show the impact of an elevated presenting polarity on verbal memory impairment, since few studies have explored the relationship between presenting polarity and neurocognition. In line with two previous studies, we also found indications (trend level significance) that patients with non-psychotic BD were more likely to have firstdegree relatives with BD, compared to patients with psychotic BD [bib_ref] Suicide attempts in bipolar I and bipolar II disorder: a review and..., Novick [/bib_ref] [bib_ref] Psychosis in bipolar disorder: phenomenology and impact on morbidity and course of..., Keck [/bib_ref]. Outside of the expected association between early age at onset and poor premorbid adjustment, there were surprisingly few group differences among age at onset subgroups.
In line with previous findings, there were clear group differences in verbal memory between BD-I and BD-II in favor of the BD-II group [bib_ref] Neuropsychological functions in patients with bipolar I and bipolar II disorder, Hsiao [/bib_ref] [bib_ref] Cognitive impairment in bipolar II disorder, Torrent [/bib_ref]. This may explain why BD-I patients have poorer general functioning than BD-II patients in spite of fewer clinical symptoms, since cognitive problems are associated with poor function [bib_ref] Neurocognitive impairment in bipolar disorder patients: functional implications, Wingo [/bib_ref]. BD-I patients had, on the other hand, better premorbid function than BD-II patients, possibly due to the earlier age at onset for the BD-II group. Due to a substantial overlap between BD-I and having a history of psychosis, group differences in verbal memory associated with history of psychosis to a large extent mirrored differences between BD-I and BD-II.
We also found support for a relationship between having psychotic BD and/or BD-I and impairments in semantic verbal fluency. Our results here are in line with previous findings of deficits in semantic verbal fluency in first-episode psychosis patients with a mania history [bib_ref] Selective deficits in semantic verbal fluency in patients with a first affective..., Kravariti [/bib_ref] and deficits in both verbal fluency and verbal memory in first-episode psychotic BD [bib_ref] Specific and generalized neuropsychological deficits: a comparison of patients with various first-episode..., Zanelli [/bib_ref]. The overlap of subgroups makes it difficult to disentangle to what extent it is the disposition to experience manic symptoms, to experience psychotic symptoms, or some common factor predisposing to both these syndromes that is associated with neurocognitive dysfunction. However, the effects of manic and/or psychotic symptomatology on cognition seem stronger than the effects of depressive symptomatology. For instance, those with depressive onsets had more depressive episodes than those with an elevated onset, but still a better performance on verbal learning and memory as well as fewer errors in general. Also, even if patients with BD-II had more current depressive symptoms, they performed better than BD-I patients on verbal learning and memory, processing speed, and verbal fluency. This is in line with a recent study that found a positive association between number of manic episodes and poorer performance on neurocognitive tests in BD-I patients, with no significant effect of number of depressive episodes [bib_ref] Effects of recurrence on the cognitive performance of patients with bipolar I..., L Opez-Jaramillo [/bib_ref]. On the other hand, these findings are equivocal, as the impact of residual depressive symptoms on cognitive domains of functioning has been demonstrated in other studies [bib_ref] The comparative clinical phenotype and long term longitudinal episode course of bipolar..., Judd [/bib_ref] [bib_ref] Clinical and neurocognitive predictors of functional outcome in bipolar euthymic patients: a..., Bonnin [/bib_ref]. The current findings seem to have clinical implications. First, as in patients with psychotic disorders [bib_ref] Cognitive training in schizophrenia: a neuroscience-based approach, Genevsky [/bib_ref] [bib_ref] Cognitive remediation for adolescents with early onset psychosis: a 1-year follow-up study, Ueland [/bib_ref] , many patients with BD have cognitive disturbances that could affect functioning and may benefit from strategies that enhance cognitive function, through cognitive remediation [bib_ref] RESEARCH: cognitive rehabilitation for bipolar disorder: an open trial for employed patients..., Deckersbach [/bib_ref] [bib_ref] Cognition and disability in bipolar disorder: lessons from schizophrenia research, Harvey [/bib_ref]. Secondly, if neurocognition is involved in the etiology and pathophysiology of the disorder, an increased understanding of this role may increase the understanding of the mechanism underlying the clinical picture and, in turn, the treatment of the disorder.
Taken together, the current findings suggest that there may be latent subgroups within the BD spectrum that to some extent encompass characteristics of several of the previously proposed subgroups; i.e., the combination of elevated presenting polarity, manic episodes and history of psychosis. These groups are characterized by impairment in neurocognitive function in particular verbal memory and semantic fluency.
# Limitations
The cross-sectional design limits the possibility to look for causal relationships. Information about onset characteristics is gathered retrospectively, with possible recall bias. Family history of psychiatric illness is based on interview with patients only. The comparison of several subgroups with repeated statistical analyses involves the risk of spurious findings, even if the main findings survive correction for multiple testing. Since this is a naturalistic study, we have not controlled use of medication, and differences in symptomatology between subgroups could be related to the use of different medications.
# Conclusions
The suggested BD subgroups show substantial overlap. At least three of the groups (BD-I, history of psychosis and elevated presenting polarity) appear to capture some common aspects of an underlying phenomenon that relates impairments in verbal memory to history of psychosis and impairments in semantic fluency to BD-I.
[table] Table 1: Clinical and neurocognitive characteristics of patients with bipolar I (BD-I) and bipolar II (BD-II) disorder [/table]
[table] Table 2: Clinical and neurocognitive characteristics of patients with bipolar disorder with and without a history of psychosis a [/table]
[table] Table 3: Clinical and neurocognitive characteristics of patients with bipolar disorder with depressive and elevated polarity of presenting episode a [/table]
[table] Table 4: Clinical and neurocognitive characteristics of patients with bipolar disorder with early, mid, and late onset a [/table]
[table] Table 5: Hierarchical regression model for verbal memory Significant results (at p < 0.05) are presented in bold. CI = confidence interval; SE = standard error. [/table]
[table] Table 6: Hierarchical regression model for semantic fluency Significant results (at p < 0.05) are presented in bold. CI = confidence interval; SE = standard error. [/table]
|
The Renin–Angiotensin System in the Tumor Microenvironment of Glioblastoma
# Introduction
Glioblastoma (GB), the most common and most aggressive primary brain cancer in humans, is classified as a WHO grade IV astrocytoma, and is characterized by microvascular proliferation and central necrosis. Primary GB arises de novo and accounts for 90% of cases with a predilection for older individuals, while secondary GB arises from low-grade astrocytoma and affects younger patients. GB has been categorized into four distinct
## Gb tumor microenvironment
The GB tumor microenvironment (TME) is highly heterogeneous and consists of cancer cells and non-cancer cells. Non-cancer cell types include immune cells, such as tumor-associated macrophages (TAMs), resident glial cells, peripheral macrophages, endothelial cells, pericytes, astrocytes, CSCs, fibroblasts, and other components such as the extracellular matrix (ECM). Given the rarity of extracranial metastasis from GB, it appears that GB development requires the unique intracerebral microenvironment inclusive of the blood-brain barrier (BBB). The TME, with emphasis on glioma-associated microglia/macrophages, pericytes, and reactive astrocytes, is increasingly recognized to play a critical role in GB development and progression . The idea that cytokines, growth factors, chemokines, inflammatory mediators, and remodeling enzymes are involved in intra-and inter-cellular communications within the TME is not novel . Additionally, constant communication between GB cells and the surrounding TMEis facilitated by extracellular vesicles that expedite bi-directional cross-talk within the TME .
Anatomically distinct regions of the TME, known as tumor niches, are thought to contain CSCs and play a fundamental role in the regulation of metabolism, immune surveillance, survival, invasion, and self-maintenance with the renin-angiotensin system (RAS) playing a critical role ]. The GB TME may consist of several distinct tumor niches including the hypoxic tumor niche, the perivascular or angiogenic tumor niche, and the vascular-invasive tumor niche. The perivascular niche contains CSCs in close juxtaposition with the abnormal angiogenic vasculature and provides a supportive environment for CSC growth, maintenance, and survival. The vasculature in the hypoxic tumor niche is either non-functional or has regressed, leading to areas of necrosis that are surrounded by rows of hypoxic palisading tumor cells . The vascular-invasive tumor niche contains tumor cells co-opted with normal blood vessels that migrate deep into the brain parenchyma .
Cancers 2021, 3 of 17 GB is highly vascular and is characterized by extensive neovascularization and pathological angiogenesis predominantly induced by vascular endothelial growth factor (VEGF), which is produced by tumor cells, CSCs, and immune cells . Other angiogenic factors, such as transforming growth factor-β 1 (TGF-β 1 ), platelet-derived growth factor-BB, and fibroblast growth factor-2, may also play a role in the pathological angiogenesis . In addition to endothelial proliferation, bone marrow-derived endothelial and pericyte progenitor cells may be recruited and incorporated into the growing vessels . There is also evidence that CSCs may be involved in neovascularization by differentiating into endothelial cells or pericytes in GB . Increased VEGF expression also fosters an immunosuppressive microenvironment that enables tumors, including GB, to evade host immune surveillance . The abnormal vasculature in GB includes dilated and leaky vessels and glomeruloid microvascular proliferation in which endothelial cells and pericytes form poorly organized vascular structures, which effectively disrupt the BBB, leading to cerebral edema. In addition, the blood-brain tumor barrier (BBTB) hinders drug delivery to the tumor .
The BBB is a highly specialized, selectively permeable barrier between the brain and the systemic blood supply that helps to maintain homeostasis of the cerebral microenvironment. The structure of the BBB includes endothelial cells with tight junctions, adherens junctions, astrocytes, pericytes, and the basement membrane . The BBB plays several fundamental roles, including supplying the brain with essential nutrients, such as oxygen and glucose, mediating the efflux of waste products, facilitating the movement of nutrients and plasma proteins, and restricting toxins into the central nervous system (CNS) . Disruption of the BBB and its tight regulation of the cerebral microenvironment leads to increased blood vessel permeability with plasma and fluid leakage into the tumor tissue causing cerebral edema and raised interstitial and intracranial pressure . The combination of abnormal vasculature in GB and the disruption of the BBB leads to impaired blood flow and reduced oxygen delivery within the tumor . Microvascular thrombosis may also occur causing occlusion of the blood vessels, further promoting intra-tumoral hypoxia, leading to pseudo-palisading necrosis . Hypoxia is also a consequence of increased oxygen diffusion distance due to the fact of tumor growth and expansion , which may, in and of itself, be a key regulator of tumor cell survival, stemness, and immune surveillance in the TME . Hypoxia also sustains tumor cell proliferation, invasiveness, and contributes to chemotherapy and radiotherapy resistance. This occurs via inhibition of free radicals, which reduces the efficacy of radiotherapy , and through upregulation of the multi-drug resistance gene, MDR1/ABCB1, which reduces chemotherapy effectiveness. Hypoxia-inducible factor-1 (HIF-1) and HIF-2 mediate the response to hypoxia on a molecular level in GB [40] and may potentially modify CSCs . The GB microenvironmental niche also consists of pseudo-palisading glioma cells that upregulate HIF proteins, inducing expression of factors, such as VEGF and interleukin 8 (IL-8), which are implicated in tumor cell survival, metabolism, invasion, and angiogenesis. The resultant cross-talk releases pro-inflammatory signals from the areas of necrosis in the hypoxic tumor niche into the surrounding TME, promoting immunosuppression, and angiogenesis.
Immune cells, including circulating monocytes, neutrophils, and myeloid-derived suppressor cells (MDSCs), are another source of angiogenic factors. In ovarian cancer, MDSCs increase CSC characteristics by increasing microRNA-101 expression, which induces the expression of stemness genes. It is interesting to speculate that MDSCs also regulate the stemness of CSCs within the GB TME via this mechanism. These cells may enter the brain as a result of breakdown of the BBB in GB and the production of tumor-derived chemokines and cytokines, contributing to the immunosuppressive GB TME. TAMs are the dominant immune cell population in GB and may include resident microglial cells and peripheral macrophages. Traditionally, TAMs have been defined as either anti-tumoral M1/Th1 (classical-activated macrophages) or protumoral M2/Th2 (alternative-activated macrophages) phenotypes. M1 macrophages foster the inflammatory response by secreting pro-inflammatory cytokines such as IL-12, tumor necrosis factor-α (TNF-α), CXCL-10, and interferon-γ (IFN-γ) and produce high levels of nitric oxide synthase to exert anti-tumor cell activity. M2 macrophages, on the other hand, play a key immunosuppressive function by secreting anti-inflammatory cytokines, such as IL-10, IL-13, and IL-4, and express abundant arginase-1, mannose receptor CD206, and scavenger receptors to promote tumor progression. The release of TGF-β by TAMs has been shown to induce matrix metalloproteinase 9 (MMP9) and, thus, increase CSC invasiveness. A more recent study has demonstrated that the TAM population is in a constant state of transition or plasticity between the two phenotypes and that M1 phenotype expression may be enhanced by TME changes or therapeutic interventions. Resident microglia are present within the brain, but it is the recruitment of peripheral macrophages to the GB TAM pool, in particular, that may mediate tumor phagocytosis with disruption of the signal regulatory protein α receptor (SIRP-α)-CD47 axis. This facilitates immune evasion because the antiphagocytic "don't eat me" surface protein CD47 is upregulated, which binds to SIRP-α on phagocytic cells to inhibit phagocytosis. However, even in the absence of macrophages, resident microglia may be transformed into effector cells of tumor cell phagocytosis, in response to anti-CD47 blockade. In models of pancreatic ductal adenocarcinoma, for example, RP-182 may selectively induce conformational switching of the mannose receptor CD206, which is expressed on the M2 TAM phenotype, ultimately reprogramming M2-like TAMs into an anti-tumor M1-like phenotype. The immunosuppressive phenotype of TAMs may be controlled by long-chain fatty acid metabolism, and chemical inhibitors targeting this metabolic pathway may block TAM polarization in vitro and tumor growth in vivo. GB-derived exosomes may reprogram M1 macrophages to M2 macrophages and condition M2 macrophages to become strongly immunosuppressive TAMs. been defined as either anti-tumoral M1/Th1 (classical-activated macrophages) or pro-tumoral M2/Th2 (alternative-activated macrophages) phenotypes. M1 macrophages foster the inflammatory response by secreting pro-inflammatory cytokines such as IL-12, tumor necrosis factor-α (TNF-α), CXCL-10, and interferon-γ (IFN-γ) and produce high levels of nitric oxide synthase to exert anti-tumor cell activity. M2 macrophages, on the other hand, play a key immunosuppressive function by secreting anti-inflammatory cytokines, such as IL-10, IL-13, and IL-4, and express abundant arginase-1, mannose receptor CD206, and scavenger receptors to promote tumor progression. The release of TGF-β by TAMs has been shown to induce matrix metalloproteinase 9 (MMP9) and, thus, increase CSC invasiveness. A more recent study has demonstrated that the TAM population is in a constant state of transition or plasticity between the two phenotypes and that M1 phenotype expression may be enhanced by TME changes or therapeutic interventions. Resident microglia are present within the brain, but it is the recruitment of peripheral macrophages to the GB TAM pool, in particular, that may mediate tumor phagocytosis with disruption of the signal regulatory protein α receptor (SIRP-α)-CD47 axis. This facilitates immune evasion because the antiphagocytic "don't eat me" surface protein CD47 is upregulated, which binds to SIRP-α on phagocytic cells to inhibit phagocytosis. However, even in the absence of macrophages, resident microglia may be transformed into effector cells of tumor cell phagocytosis, in response to anti-CD47 blockade. In models of pancreatic ductal adenocarcinoma, for example, RP-182 may selectively induce conformational switching of the mannose receptor CD206, which is expressed on the M2 TAM phenotype, ultimately reprogramming M2-like TAMs into an anti-tumor M1-like phenotype. The immunosuppressive phenotype of TAMs may be controlled by long-chain fatty acid metabolism, and chemical inhibitors targeting this metabolic pathway may block TAM polarization in vitro and tumor growth in vivo. GB-derived exosomes may reprogram M1 macrophages to M2 macrophages and condition M2 macrophages to become strongly immunosuppressive TAMs. end-product of the paracrine RAS, activates ATII receptor 1 (AT 1 R) leading to increased tumor cell proliferation, oxidative stress, hypoxia and angiogenesis, and inflammation-the hallmarks of cancer. This contributes to an inflammatory TME by increasing the number of inflammatory cells, partly by increasing the number of NADPH complexes, leading to tumor cell proliferation, DNA damage from oxidative stress, and release of growth factors. AT 1 R also activates phosphatidylinositol signaling, which increases cytosolic Ca 2+ to promote mitogenesis. Hypoxia increases paracrine RAS activity by upregulating angiotensin-converting enzyme (ACE) and the expression of hypoxia-inducible factor 1α (HIF-1α) and HIF-2α, which increase tumor progression and treatment resistance. HIF-1α, HIF-2α, and hypoxia increase the expression of vascular endothelial growth factor (VEGF) which increases angiogenesis. AT 1 R, via MAPK-STAT3 signaling, contributes to a cytokine release that leads to CSC renewal. C-X-C chemokine receptor type 4 (CXCR4) promotes tumor cell migration and invasion. AT 1 R signaling and the prorenin receptor, which act in a feedback loop with Wnt/β-catenin, increase Wnt signaling which promotes CSC stemness by upregulating stemness-associated markers. Myeloid-derived suppressor cells (MDSCs) promote CSC characteristics by increasing microRNA-101 expression that induces expression of stemness-related genes in CSCs. The Ang(1-7)/MasR axis opposes the ACE/ATII/AT 1 R axis. Cathepsins B, D, and G act as bypass loops for the RAS. Under the influence of the TME, polarization of tumor-associated macrophages (TAMs)-immune cells that are located within the TME-changes from the M1 to M2 phenotype. M2 TAMs induce the proliferation of CSCs via interleukin 6 (IL-6)-induced activation of STAT3, leading to cytokine release and positive feedback contributing to CSC renewal. Glioblastoma CSCs secrete Wnt-induced signaling protein 1 (WISP1), which facilitates a pro-tumor TME by promoting the survival of CSCs and M2 TAMs, and also promotes CSC maintenance. Abbreviations: ATI, angiotensin I; AT 2 R, ATII receptor 2; Ang(1-7), angiotensin 1-7; ATIII, angiotensin III; MAPK, mitogen-activated protein kinase.
## Glioblastoma cancer stem cells
The CSC concept proposes that a small distinct population of cells within a tumor with self-renewal capability are responsible for driving tumorigenesis. These CSCs may be defined as stem cell-like cells within a tumor that also have the capacity for proliferation and multi-potency. This may be regarded as a functional definition insofar as CSCs may be characterized through the generation of serially transplantable tumors that faithfully recapitulate the parent tumor. There is marked intra-and inter-tumoral heterogeneity including, differing numbers of highly tumorigenic CSCs. Such heterogeneity may be best explained by a combination of different models of cancer, including the stochastic model (also known as the clonal evolution model), the CSC concept of cancer (also known as the hierarchical model of cancer), and the concept of plasticity.
The traditional model of cancer is predicated on the stochastic model of carcinogenesis which proposes that cancer cells are derived from normal cells that acquire genetic and/or epigenetic mutations resulting in typically unidirectional transitions from benign to malignant cells. These malignant tumor cells have unrestricted division capacities and their high mutation rates increase the likelihood of successive generations of cloned cells being adapted to the selection pressures of the tumor site. However, the stochastic model does not fully account for all aspects of cancer biology including tumor recurrence following treatment.
In contrast, the CSC concept of cancer proposes that CSCs contribute to carcinogenesis, invasion, metastasis, therapy resistance, and recurrence. CSCs divide asymmetrically into non-tumorigenic cancer cells, which form the bulk of a tumor, and identical highly tumorigenic but less abundant CSCs, which sit at the apex of the cellular hierarchy. CSCs have been postulated to originate from non-malignant stem cells or progenitor cellsor dedifferentiated cancer cells. The overlap between the stochastic model and the CSC concept may be explained by the concept of cellular plasticity whereby cancer cells may reversibly transition between stem-like and non-stem-like cell states. This process of transition may be driven by embryonic stem cell (ESC)-associated regulatory networks and may be affected by the dynamic TME including the CSC niche. Moreover, certain cancer cells may de-differentiate and re-enter the CSC pool, thus regaining the capacity for tumorigenesis and clonal expansion.
CSCs have been found in many different cancer types, including myeloid leukemia, pancreatic cancer, breast cancer, oral cavity squamous cell carcinoma (SCC), primaryand metastaticcutaneous SCC, primaryand metastaticcolon adenocarcinoma, metastatic malignant melanoma, and GB. The aggressive nature of GB and its resistance to conventional therapy has been attributed to the presence of CSCsthat were first postulated in human brain tumors, identified by their expression of the neural stem cell surface marker CD133. Stem-like neural precursor cells responsible for the growth and recurrence in serial transplantations were identified in GB. The presence of such quiescent CSCs is well-supported in the literature and the interaction of such cells with the ECM and TME factors, including TGF-β and hypoxia, may contribute to their resistance to conventional therapy. There is evidence that CSCs may be stimulated to differentiate into endothelial cells by activating Notch1 signalingand may be associated with induction of cytokines, MMPs, and adhesion proteins in the TME.
A crucial function of stem cells is self-renewal, for which the Notch, Sonic hedgehog, and Wnt signaling pathways may be essential. GB expresses a number of stemness-associated markers including cell surface markers (CD133, CD15, A2B5, and L1CAM), cytoskeletal proteins (nestin), transcription factors (SOX2, NANOG, and OCT4), post-transcriptional factors (Musashi1), and polycomb transcriptional suppressors (Bmi1 and Ezh2). There is also evidence of plasticity and bi-directional interconversion between CSCs and cancer cells. In a landmark study, pluripotent stem cells were formed from reprogrammed mouse embryonic and adult fibroblasts by the addition of transcription factors OCT4, SOX2, c-MYC, and KLF4. These factors, in addition to NANOG, which are expressed by ESCs, have been identified in GB. The capacity of GB cells for perpetual self-renewal may rely on the contribution from transcription factors such as OCT4 and SOX2. SOX2 is highly expressed in GBand may play a key role in maintaining plasticity for bi-directional cellular conversion in GB. Moreover, silencing of SOX2 inhibits tumor proliferation in GBand, thus, it may be a potential therapeutic target in the treatment of GB. Another potential therapeutic target involves the JAK-STAT3 signaling pathway which is also associated with the self-renewal capacity of GB. Inhibition of this pathway may impede the migratory and invasive potential of GB by decreasing activation of the transcription factor STAT3 and, thus, reducing the levels of MMPs and associated invadopodia activity. In addition, STAT3 binding to the Notch1 promoter inhibits this signaling pathway and may impede the maintenance of glioma stem-like cells while reducing the expression of glioma stem cell markers CD133, SOX2, and nestin.
## The renin-angiotensin system and convergent signaling pathways in glioblastoma
The RAS has been proposed to play an important role in the TME [19] in various cancer types, including lung cancer, through its effect on tumor cells, non-malignant cells, hypoxia, angiogenesis, and the inflammatory response. The RAS is a complex physiological system and has a multitude of interactions with many different convergent signaling pathways that operate in carcinogenesis, some of which lie outside the scope of this article.
Classically, the RAS regulates blood pressure and electrolyte and fluid homeostasis involving primarily the renal, cardiovascular, and endocrine systems. The RAS pathway is composed of multiple steps culminating in the formation of the main effector hormone, angiotensin II (ATII). Activity of this key homeostatic system in the CNS is well documented. In this review article, RAS inhibition broadly refers to inhibition of any of the components of the RAS, reducing its downstream effects.
Angiotensinogen, primarily synthesized in the liver by hepatocytes, is cleaved by renin, to form angiotensin I (AT1). Angiotensinogen is synthesized and secreted by astrocytes and is converted to several neuroactive peptides. Angiotensinogen is also produced within neurons, which can secrete or retain it intracellularly. These neuroactive peptides bind their respective receptors within the local microenvironment to induce receptor signaling by different cell types. Renin is physiologically derived from the juxtaglomerular apparatus in the kidneys and its release is tightly regulated by macula densa and local baroreceptors. Renin is formed by the binding of prorenin to the prorenin receptor (PRR)and is also catalyzed by enzymes such as cathepsins B, D, and G. ATI is converted to ATII by angiotensin-converting enzyme (ACE), also known as ACE1, which is primarily found in the lungs. ATII binds to ATII receptor 1 (AT 1 R) and ATII receptor 2 (AT 2 R). ATII binding to AT 1 R causes MAPK-STAT3 activationand phosphatidylinositol signaling, which increases cytosolic Ca 2+ and effects mitogenesis. AT 1 R signaling increases RAS activity in the TME, and the formation of NF-κB and TGF-β 1 which promotes cellular proliferation, inflammation, and angiogenesis. AT 2 R activation by ATII inhibits cellular growth and enhances apoptosis. ATII can be further converted into angiotensin III (ATIII), and then angiotensin IV (ATIV) by aminopeptidase-A (AP-A) and aminopeptidase-N (AP-N), respectively. ATIV binds to ATII receptor 4 (AT 4 R), and in high concentrations, may bind to AT 1 R. Angiotensin (1-7)is produced by the cleavage of either ATI by neutral endopeptidase (NEP) or ATII by ACE2, an isoform of ACE. Ang(1-7) binds to Mas receptors (MasRs). ATI may also be cleaved by ACE2 to form Ang(1-9), which can be cleaved by ACE1 and is converted to Ang(1-7), which in addition to binding to MasRs, can also bind to AT 2 R with low affinity, and Mas-related-G protein coupled receptors (MrgDs). MrgDs are a recently discovered component of the RAS, and their role in the GB TME is yet to be defined. Lastly, the primary ligand for MrgDs is almandine, an Ang(1-7) analog formed by decarboxylation of Ang(1-7) .
Key components of the RAS are also activated in CNS diseases. Renin, and its precursor prorenin, are expressed variably in neurons, astrocytes, oligodendrocytes, and microglia in different regions of the brain. PRR is widely distributed in different organs throughout the body including the brain, eyes, and immune system. ACE1 is expressed in areas of the brain involved in blood pressure control and homeostasis including the choroid plexus, organum vasculosum of the lamina terminalis, subfornical organ, and area postrema. ACE2 is found in the endothelium of the brain in various regions including the cortex and brainstem. ACE2 contributes to the neuroprotective ACE2/Ang(1-7)MasR signaling axis by converting ATII to Ang(1-7) which is a ligand for MasR.
The RAS, as a constituent of the TME, is involved in several hallmarks of cancer, including angiogenesis, hypoxia, and tumor cell proliferation. Components of the RAS are expressed in different types of cancer including colon adenocarcinomaand malignant melanoma. RAS components are also expressed by CSCs in oral cavity SCC, renal clear cell carcinoma, primary, and metastatic, cutaneous SCC, metastatic colon adenocarcinoma, metastatic malignant melanoma, and GB. In GB, PRR, AT 1 R, and AT 2 R are co-expressed with stemness-associated markers. PRR is highly expressed in GB compared with lower-grade gliomas; this higher expression of PRR in higher-grade glioma is notable as the Wnt/β-catenin signaling pathway is implicated in the self-renewal of stem cells.
The Wnt/β-catenin signaling pathway, which sits downstream of the RAS, is implicated in tumor initiation in several cancer types. In brief, this pathway results in active β-catenin translocating into the nucleus, upregulating the expression of oncogenes such as c-Myc, AXIN2, and CCND1. PRR is a component of the Wnt receptor complex and acts as an adapter between vacuolar H + -adenosine triphosphate (V-ATPase) and low-density lipoprotein receptor-related protein 6. V-ATPase, a proton pump, is essential for cellular acidification and is involved in the mechanism for β-catenin activation. This process facilitates binding of Wnts to their respective Wnt receptor complex. Further, PRR promotes brain cancers via the Wnt/β-catenin signaling pathway, and in addition to being a membrane receptor, exists in the cytoplasm and increases the protein expression of Wnt2 within glioma cells. This evidence underscores the PRR as a potential oncoprotein via Wnt/β-catenin pathway-related carcinogenesis, which influences cell stemness, tumorigenesis, and cellular proliferation. Renin is expressed in GB and may contribute to the mechanisms of neovascularization in GB. Furthermore, downregulation of the Ang(1-7)/MAS signaling axis by podocalyxin results in enhanced GB cell invasion and proliferation. Finally, bypass loops of the RAS involving various cathepsins that may also contribute to the proliferative activity in GB, for example, cathepsin G coverts ATI to AII and from AGT directly to ATII, which binds to AT 1 R, to promote cancer progression. GB CSCs have been shown to secrete Wnt-induced signaling protein 1 (WISP1) that promotes the survival of both the CSCs and M2 TAMs to promote a pro-TME. STAT3 activationand phosphatidylinositol signaling, which increases cytosolic Ca2 + and effects mitogenesis. AT1R signaling increases RAS activity in the TME, and the formation of NF-κB and TGF-β1 which promotes cellular proliferation, inflammation, and angiogenesis. AT2R activation by ATII inhibits cellular growth and enhances apoptosis. ATII can be further converted into angiotensin III (ATIII), and then angiotensin IV (ATIV) by aminopeptidase-A (AP-A) and aminopeptidase-N (AP-N), respectively. ATIV binds to ATII receptor 4 (AT4R), and in high concentrations, may bind to AT1R. Angiotensin (1-7)is produced by the cleavage of either ATI by neutral endopeptidase (NEP) or ATII by ACE2, an isoform of ACE. Ang(1-7) binds to Mas receptors (MasRs). ATI may also be cleaved by ACE2 to form Ang, which can be cleaved by ACE1 and is converted to, which in addition to binding to MasRs, can also bind to AT2R with low affinity, and Mas-related-G protein coupled receptors (MrgDs). MrgDs are a recently discovered component of the RAS, and their role in the GB TME is yet to be defined. Lastly, the primary ligand for MrgDs is almandine, an Ang(1-7) analog formed by decarboxylation of[102] .
## Figure 2.
A schema showing the effect of the renin-angiotensin system (RAS) and its convergent signaling pathways on the tumor microenvironment to influence cellular proliferation, invasiveness, and cell survival in cancer development. The RAS interacts with downstream pathways, such as the Ras/RAF/MEK/ERK (light blue) pathway and the PI3K/AKT/mTOR (dark blue) pathway, and the upstream Wnt/β-catenin pathway (intermediate blue) that influence cellular proliferation, migration, inhibition of apoptosis, migration, and invasion (see text). PRR, pro-renin receptor; LRP6, low-density lipoprotein receptor-related protein; Fzd, frizzled receptor; Cath G, cathepsin G; Cath B, cathepsin B; Cath D, cathepsin D; ACE1, angiotensin-converting enzyme 1; ACE2, angiotensin-converting enzyme 2; ADP, adenosine diphosphate; AGT, angiotensinogen; ATP, adenosine triphosphate; Ang(1-7), angiotensin (1-7); Ang(1-9), angiotensin (1-9); AP-A, aminopeptidase-A; NEP, neutral endopeptidase; AP-N, aminopeptidase-N; ATI, angiotensin I; ATII, angiotensin II; ATIII, angiotensin III; ATIV, angiotensin IV; AT 1 R, angiotensin II receptor 1; AT 2 R, angiotensin II receptor 2; AT 4 R, angiotensin II receptor 4; MrgD, Mas-related-G protein coupled receptor; MasR, Mas receptor; mTOR, mammalian target of rapamycin; NF-κB, nuclear factor kappa B; TGF-β 1 , transforming growth factor-β 1 ; V-ATPase, vacuolar H + -adenosine triphosphate.
Other related signaling pathways, such as the PI3K/AKT/mammalian target of rapamycin (mTOR) and Ras/RAF/MEK/ERK pathways within the GB TME, downstream to the RAS, are activated via AT 1 R and PRR signal transduction. MAPK/ERK signaling is activated upon binding of renin or prorenin to PRR, and this upregulates ERK1/2 in various cell types including neurons. ERK1/2 activation induces TGF-β 1 formation and Cancers 2021, 13, 4004 9 of 17 cellular proliferation, both of which influence cancer development. Supporting this is the fact that silencing of PRR downregulates expression of ERK1/2, AKT, and NF-κB. Additionally, PRR activation leads to the production of reactive oxygen species, which activates both the PI3K/AKT/mTOR and Ras/RAF/MEK/ERK pathways . It is interesting to speculate that both pathways operate in conjunction with the RAS and Wnt/β-catenin to influence proliferation, survival, stemness, and invasiveness of CSCs within the GB TME.
The use of RAS inhibitors (RASis) in the treatment of cancer may mitigate the cytotoxic treatment-related adverse effects experienced by cancer patients to improve their overall quality of life. A meta-analysis of 17 observational studies by Shen et al.show RASis are associated with a reduced risk of cancer. A prospective populationbased study also shows long-term (>3 years) administration of RASis is associated with a decreased risk of cancer in patients with a DD genotype, which is associated with high levels of ACE and, thus, increased RAS activity. This is relevant as increased levels of ATII caused by elevated RAS activity promotes cancer progression by its actions on AT 1 R. Other epidemiological studies have shown a protective benefit of RASis against colorectal cancerand an overall reduced risk of cancer. RASis have also been shown to improve the overall survival of patients with aggressive non-metastatic pancreatic ductal adenocarcinoma. Although current data remain inconclusive, RASis appear to be broadly protective against cancer.
A retrospective study analyzing clinical data from 810 patients enrolled in two large multicenter studies investigating the role of two drugs targeting the RAS combined with statins in GB, shows no benefit in overall survival. A recent trial on repurposing multiple drugs in combination with temozolomide, including two drugs that affect the RAS (i.e., captopril and celecoxib) for patients with GB, observed maintenance of good quality of life. Captopril, an ACE inhibitor, and celecoxib, which inhibits cyclocoxygenase-2, reduce RAS activity . In addition, RASis, in combination with bevacizumab, improve survival in patients with GB, although there is no overall survival benefit of this VEGF inhibitor as a monotherapy for de novo or recurrent GB. PRR may be a critical biomarker and a therapeutic target for the treatment of GB with its connections to V-ATPase function, and the Wnt/β-catenin, MAPK/ERK, and PI3K/AKT/mTOR pathways. Several other steps of the RAS pathway can potentially be targeted. The effects of a novel approach, targeting the RAS, its bypass loops, and converging pathways simultaneously using multiple repurposed drugs on the quality of life and progression-free survival in GB patients are currently being investigated in a clinical trial. Therapeutic options may be facilitated by augmenting the compensatory mechanisms of the RAS.
## Recent developments
Recent technological breakthroughs in generating human cerebral organoidsfrom pluripotent cells, combined with genetic engineering, mass spectroscopic proteomics, and next generation gene sequencing tools, allow more detailed investigation into the GB TME, and the role of the RAS in this niche. Cerebral organoids have been shown to more faithfully recapitulate the temporal and spatial aspects of the developing brain. Vascularized cerebral organoids have been developed by utilizing ectopic expression of human ETS variant 2 in engineered ESCs to form a vascular-like network in organoids akin to endothelial cells. In addition, VEGF has been used to induce blood vessel-like structures in cerebral organoids expressing markers associated with the BBB, namely, CD31 and claudin-5. In addition, human umbilical vein endothelial cells have been used to develop cerebral organoids with a well-developed tubular vascular structure. In another notable development, choroid plexus-like organoids modeled cerebrospinal fluid production with a selective barrier akin to the BBB, which may be used to model the BBTB in the GB TME. Using RNA sequencing, moreover, GB cerebral organoid models have been shown to best mimic the cellular states and plasticity found in the GB TME compared to gliospheres, tumor organoids, and orthotopic patient-derived xenografts.
# Conclusions
Despite intensive research into the biology and treatment of GB, the prognosis of patients with GB remains dismal. Understanding the heterogeneity of the tumor-host microenvironment in GB, the role of RAS and CSCs, and mapping salient interactions on a cellular level employing techniques, such as single-cell RNA sequencing, may lead to the discovery of potential therapeutic targets. Cerebral and GB organoids represent an exciting yet relatively cost-effective way to delineate relevant signaling pathways within the GB TME, including the RAS, and provide models for developing and testing drug screening and therapeutic targets including RASis. |
Assembly of supramolecular DNA complexes containing both G-quadruplexes and i-motifs by enhancing the G-repeat-bearing capacity of i-motifs
The single-step assembly of supramolecular complexes containing both i-motifs and G-quadruplexes (G4s) is demonstrated. This can be achieved because the formation of four-stranded i-motifs appears to be little affected by certain terminal residues: a fivecytosine tetrameric i-motif can bear ten-base flanking residues. However, things become complex when different lengths of guanine-repeats are added at the 3 or 5 ends of the cytosine-repeats. Here, a series of oligomers d(XG i XC 5 X) and d(XC 5 XG i X) (X = A, T or none; i < 5) are designed to study the impact of Grepeats on the formation of tetrameric i-motifs. Our data demonstrate that tetramolecular i-motif structure can tolerate specific flanking G-repeats. Assemblies of these oligonucleotides are polymorphic, but may be controlled by solution pH and counter ion species. Importantly, we find that the sequences d(TG i AC 5 ) can form the tetrameric i-motif in large quantities. This leads to the design of two oligonucleotides d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ) that selfassemble to form quadruplex supramolecules under certain conditions. d(TG 4 AC 7 ) forms supramolecules under acidic conditions in the presence of K + that are mainly V-shaped or ring-like containing parallel G4s and antiparallel i-motifs. d(TG Br GG Br GAC 7 ) forms long linear quadruplex wires under acidic conditions in the presence of Na + that consist of both antiparallel G4s and i-motifs.
# Introduction
The G-rich oligonucleotide sequences containing tandem G-stretches can fold into tetra-stranded DNA structures called G-quadruplexes (G4s) that may play an important role in biological processes such as immunoglobu-lin gene rearrangements, promoter activation and telomere maintenance [bib_ref] G-quadruplex nucleic acids and human disease, Wu [/bib_ref]. Many factors such as sequence, cationic species and concentration, small molecule ligands, molecular crowding and more affect the stability of G4s and induce the formation of polymorphs [bib_ref] G-quadruplex DNA assemblies: loop length, cation identity, and multimer formation, Smargiasso [/bib_ref] [bib_ref] Stability and molecular recognition of quadruplexes with different loop length in the..., Arora [/bib_ref]. Furthermore, the existence of complementary C-rich strands has significant impact on the G4 DNA formation [bib_ref] Human telomeric DNA: G-quadruplex, i-motif and Watson-Crick double helix, Phan [/bib_ref] and under slightly acidic conditions such C-rich sequences can form their own tetrastranded non-canonical secondary structure known as an i-motif [bib_ref] The i-motif in nucleic acids, Guéron [/bib_ref]. Such C-rich strands with the potential to fabricate i-motif scaffolds are frequent in the human genome [bib_ref] Formation of the G-quadruplex and i-motif structures in retinoblastoma susceptibility genes (Rb), Xu [/bib_ref] and may also have crucial importance in biological functions [bib_ref] Fundamental aspects of the nucleic acid i-motif structures, Benabou [/bib_ref].
In a G4, Hoogsteen hydrogen bonds connect guanines within quartets, which stack with mono-or divalent cations in the central channel to enhance its stability (Scheme 1A and B) [bib_ref] G-quartet structures in telomeric DNA, Williamson [/bib_ref] [bib_ref] A lead-filled G-quadruplex: insight into the G-quartet's selectivity for Pb 2+ over..., Kotch [/bib_ref]. For unimolecular and bimolecular G4s the paths of various loops connecting stacked G's above grooves must also be considered. In contrast, an i-motif is made up of two parallel duplexes held together by hemiprotonated cytosine-cytosine (C- CH + ) pairs (Scheme 1C and D) [bib_ref] Investigation of the energetics of C-H···O hydrogen bonds in the DNA i-motif..., Leroy [/bib_ref]. These duplexes are alternatively intercalated in an antiparallel orientation to form a stable and long-lived four-stranded nucleic acid secondary structure that is stabilized byand electrostatic interactions [bib_ref] Fundamental aspects of the nucleic acid i-motif structures, Benabou [/bib_ref] [bib_ref] The formation pathway of i-motif tetramers, Leroy [/bib_ref]. Compared to antiparallel B-form Watson-Crick (WC) duplex structures, quadruplex structures may be more flexible and functional due to their richer base stacking and abundance of ligand binding sites [bib_ref] Four-stranded nucleic acids: structure, function and targeting of G-quadruplexes, Huppert [/bib_ref].
The simple fact that sequential runs of G's or C's occur more frequently in the genome than expected by random chance, indicates that they are likely to have biological significance [bib_ref] The probability of occurrence of oligomer motifs in the human genome and..., Macleod [/bib_ref]. Previous studies demonstrate that poly[d(G + C)] oligonucleotides can easily form different types of duplexes, for example, B-form, A-form or Z-form [bib_ref] High-salt d(CpGpCpG), a left-handed Z DNA double helix, Drew [/bib_ref] [bib_ref] A short GC-rich palindrome of human mannose receptor gene coding region displays..., Bansal [/bib_ref] and sometime even quadruplexes [bib_ref] Duplex to quadruplex equilibrium of the self-complementary oligonucleotide d (GGGGCCCC), Deng [/bib_ref] [bib_ref] Guanine tetraplex formation by short DNA fragments containing runs of guanine and..., Penázová [/bib_ref] [bib_ref] Quadruplex formation by both G-rich and C-rich DNA strands of the C9orf72..., Zamiri [/bib_ref]. In addition, studies have revealed that the C-runs or cytosine residues in (G + C)-rich oligonucleotides significantly affect the formation [bib_ref] Conserved guanine-guanine stacking in tetraplex and duplex DNA, Kypr [/bib_ref] and stability of G4s [bib_ref] Cation-dependent transition between the quadruplex and Watson-Crick hairpin forms of d(CGCG 3..., Hardin [/bib_ref] [bib_ref] Cytosine-cytosine + base pairing stabilizes DNA quadruplexes and cytosine methylation greatly enhances..., Hardin [/bib_ref]. Mao and his research Scheme 1. Schematic descriptions of (A) a G-quartet linked by Hoogsteen hydrogen bonds, (B) the chelation cage formed by eight O 6 atoms of guanines and K + ; (C) a C- CH + base pair, (D) two pairs of maximally intercalated C- CH + base pairs. group show that for certain double-stranded sequences, the formation of G4s and i-motifs is mutually exclusive even under a favorable condition, which is mainly governed by steric hindrance in the double helix structure [bib_ref] G-quadruplex and i-motif are mutually exclusive in ILPR double-stranded DNA, Dhakal [/bib_ref] [bib_ref] Mutually exclusive formation of G-quadruplex and i-motif is a general phenomenon governed..., Cui [/bib_ref]. Earlier studies also show that tetra-stranded i-motif structures formed by five consecutive cytosines have comparable stability to ten-base-pair-long duplexes [bib_ref] DNA pillars constructed from an i-motif stem and duplex branches, Yang [/bib_ref]. However, as yet, there are no systematic investigations into the influence of guanine residues or complementary G-runs on the formation of i-motif structures. Here, we show that in the right context this i-motif remains stable in the presence of a run of three or more G's; thus, intermolecular i-motif structures can outcompete intermolecular WC pairs with the right sequence and conditions. DNA scaffolds with specific structural features and remarkable molecular recognition properties are one of the most promising materials for building nanoscale structures. Two principles for DNA macromolecular design have emerged over the past three decades: first, DNA secondary structures can serve as monomeric units; second, these monomers must have flexibility in order to further assemble into larger structures [bib_ref] Design and self-assembly of two-dimensional DNA crystals, Winfree [/bib_ref] [bib_ref] Challenges and opportunities for structural DNA nanotechnology, Pinheiro [/bib_ref]. The principles have allowed the assemblage of DNA supramolecular complexes into precise topological structures [bib_ref] Synthesis from DNA of a molecule with the connectivity of a cube, Chen [/bib_ref] [bib_ref] Metal-nucleic acid cages, Yang [/bib_ref] and crossover DNA tiles [bib_ref] Algorithmic self-assembly of DNA Sierpinski triangles, Rothemund [/bib_ref] [bib_ref] Surface-mediated DNA self-assembly, Sun [/bib_ref]. We believe that i-motif assemblies should be excellent construction modules to facilitate supramolecular DNA complexes due to their ability to assemble specifically and the flexibility of the motif [bib_ref] DNA nanotechnology based on i-motif structures, Dong [/bib_ref]. For example, oligonucleotide hybrid supramolecules composed of parallel or antiparallel duplexes and tetrameric i-motifs have been reported by . Others have shown that complementary DNA fragments [bib_ref] DNA pillars constructed from an i-motif stem and duplex branches, Yang [/bib_ref] and C-rich repeats [bib_ref] association into supra molecular i-motif structures, Laisné [/bib_ref] [bib_ref] Junctions between i-motif tetramers in supramolecular structures, Guittet [/bib_ref] located at the terminal of tetra-stranded i-motifs also act as flexible building blocks to allow the assembly supramolecular DNA structures via the formation of double helices and tetra-stranded i-motif structures. However, none of these systems have both C's and G's in the same oligonucleotide that each contribute to distinct tetraplexes. There is a special problem that G's and C's present in the same oligonucleotide may complement each other if the strand folds on itself or contribute to the formation of competing duplexes between strands. We postulate that with appropriate restrictions tetra-stranded i-motif structures can bear terminal Grepeats capable of folding into four-stranded G4s that contribute to the assembly of DNA nanowires containing both G4s and i-motifs.
In this work, short DNA strands d(TG i TC 5 T) and d(TC 5 TG i T) (i < 5) are explored as the primary sequences to evaluate the above postulates. Thymines are used as the 'spacers' because they stabilize i-motif structures either in the middle or at the end of C-rich strands [bib_ref] Formation and dissociation of the interstrand i-motif by the sequences d(X n..., Cao [/bib_ref] [bib_ref] an i-motif tetramer with intercalated T•T pairs, Canalia [/bib_ref] , and they are less prone to stacking or pairing with guanines [bib_ref] Design of a G-quadruplex topology through glycosidic bond angles, Webba Da Silva [/bib_ref]. The rational investigation of C-runs with spacer and flanking residues has directed us to model i-motifs that can bear longer G-runs for the formation of G4s. To our knowledge, this is the first systematic study of DNA sequences composed of C-and G-runs and their assembly simultaneously into both i-motifs and G4s in a designed fashion.
# Materials and methods
## Materials and sample preparation
The oligonucleotides (OPC grade for ordinary sequences and HPLC grade for the modified sequence) were synthesized by TaKaRa Biotechnology Co., Ltd. (Dalian, China) and used as received. The DNA fragments studied here are listed in [fig_ref] Table 1: Sequences of oligodeoxynucleotides studied here [/fig_ref]. Stock solutions of the oligonucleotides were prepared by directly dissolving the lyophilized powder in Milli-Q water to approximately 1 mM and were stored at −20 - C. Accurate concentration of the oligonucleotides were obtained from their UV absorbances at 260 nm at 90 - C using molar absorptivities tabulated at the website (http:// scitools.idtdna.com/scitools/Applications/oligoAnalyzer).
For the supramolecular DNA assembly, 40-100 M of each oligomer was placed in pH 4.5 (acetic acid-acetate) or pH 9.0 (acetate-ammonia-water) buffer solution with different counter ions (50-100 mM NH 4 + , Li + , Na + or K + ions). Samples were heated in a 90 - C water bath for 15 min and slowly cooled to room temperature (about 10 h), followed by equilibration at 4 - C for more than 4 days.
## Mass spectrometry
Mass spectrometric analysis was performed in ESI negativeion mode on an ESI-Q-TOF (micrOTOF-Q II, Bruker, Breman, Germany) mass spectrometer controlled by Bruker ESI Compass Data Analysis 4.0 software. This system can measure m/z in the range 50-3000. The samples were infused directly into ion source at a flow rate of 3 l min −1 . Optimal soft ionization conditions for quadruplex structures were found using the oligomer d(G 4 T 4 G 4 ) to tune the instrument to produce abundant [(dG 4 T 4 G 4 ) 2 + 3NH 4 ] 5− ions as described previously [bib_ref] Formation and dissociation of the interstrand i-motif by the sequences d(X n..., Cao [/bib_ref] [bib_ref] Ammonium ion binding to DNA G-quadruplexes: do electrospray mass spectra faithfully reflect..., Balthasart [/bib_ref]. These conditions also produce good spectra for i-motifs [bib_ref] Formation and dissociation of the interstrand i-motif by the sequences d(X n..., Cao [/bib_ref]. Before the MS analysis samples were diluted with an equal volume of methanol solution 60% in water in order to obtain a good spray [bib_ref] Special feature: tutorial-meeting report-interaction between antitumor drugs and a double-stranded oligonucleotide studied..., Gabelica [/bib_ref]. The data collection time was 0.6 min for each spectrum. For MS/MS experiments, precursor ions were selected and isolated in the quadrupole and then fragmented in the collision cell. The in-source collision induced dissociation (ISCID) energy was set at 50 eV. The collision energy was varied over the range of 10-30 eV.
## Ms ion distribution analysis
The gas phase ion abundances and theoretical isotope distributions were obtained by using Bruker ESI Compass Data Analysis 4.0 software. The percentage (TP) of each ion species is calculated using Equation (1):
[formula] T P = I nM × n I M + I 2M × 2 + I 3M × 3 + I 4M × 4 × 100% (1) [/formula]
where the numerator is the sum of all strands in one ion species, and the denominator is the sum of all strands in all ions. Each I is the intensity of detected ion in different charge states. For simplicity, the ion species such as fragment ions in the gas phase and hydrolyzed ions in solution were not counted here due to their low abundance or low influence on the overall results. Additionally, when a sequence generated a series of [4M + yNH 4 ] x− ions (y = 0 ∼ 3), only the strongest ion peak was counted. When the peaks from monomers and dimers overlap with the tetramer, their intensities are subtracted from the tetramer intensities based on the calculated isotopic profiles. Similarly, isotopic profiles were also used to estimate intensities of peaks with m/z out of range.
## Cd spectroscopy
CD spectra were recorded on a J-810 CD spectrometer (JASCO, Tokyo, Japan) using 0.2 mm path length Hellma cell at the room temperature. Each sample of 40 M DNA strand was dissolved in 50 mM NH 4 OAc or LiOAc (pH 4.5 or 9.0). Spectra were recorded from 320 to 200 nm as the averages of three scans. Each trace was measured at 100 nm/min of scanning speed, with 2 s response time, 1 nm data pitch and 1 nm bandwidth. The background spectra corresponding to the buffer alone were subtracted from all DNA spectra. Characteristic peaks of various nucleic acid structures are listed in [fig_ref] Table 1: Sequences of oligodeoxynucleotides studied here [/fig_ref].
## Uv absorption spectrophotometry
UV absorbance versus temperature melting profiles were obtained using a UV-2550 spectrophotometer (SHIMADZU, Kyoto, Japan) equipped with a S-1700 temperature controller and measured at 260, 265 and 295 nm, respectively, according to previous reports [bib_ref] Following G-quartet formation by UV-spectroscopy, Mergny [/bib_ref] [bib_ref] Analysis of thermal melting curves, Mergny [/bib_ref]. The samples with final DNA concentrations of 40 M dissolved in 50 mM LiOAc at pH 4.5 or 9.0 were used for UV experiments. In each step, the temperature was increased by 0.5 - C and equilibrated for 0.5 min before recording the absorbance. The absorption cell was sealed to avoid solvent evaporation, and the gas bubbles generated in heating process were removed by stirring.
## Native gel electrophoresis
Gel electrophoresis was run with an 8% poly/acrylamide gel at 4 - C controlled by a condenser circulating water system. 2-(N-morpholino) ethanesulfonic acid monohydrate (MES, pH 4.5, 50 mM) buffer was used to prepare the gel as well as the running buffer, which was further supplemented with 100 mM KOAc. Each sample used for Native Gel Electrophoresis contained 100 M DNA strand and 100 mM counter ions (either Li + , Na + or K + ions) at pH 4.5 or 9.0. The electrophoresis was performed for 3 h at 150 V. Results were visualized by Ultrapower™ visible light transilluminator (Bioteke, Beijing, China) after using Gelgreen DNA staining agent (Biotium) in 100 mM NaCl. The gel image was enhanced using a grayness transform by using Adobe Photoshop CS6 software to reduce background interference.
## Atomic force microscopy
Atomic force microscopy (AFM) was performed on mica surfaces that bind nucleic acids with bridging magnesium ions [bib_ref] Optical properties of guanine nanowires: experimental and theoretical study, Changenet-Barret [/bib_ref]. Here, the DNA samples (100 M of each DNA) were annealed in different buffers containing 100 mM cations at 4 - C for 1 week. Then aliquots were diluted with 2 mM MgCl 2 aqueous solution to give a 20 l analyte containing 1.5 M DNA. The analytes were incubated on freshly cleaved mica plates for 5-8 min. After that, excess fluid was removed. Then the mica plates were washed carefully and repeatedly with Milli-Q water, and left to air-dry, in a clean container to prevent dust settling. AFM analysis was in tapping mode with a Nanoscope IIIa scanning probe microscope (Bruker) from Digital Instruments utilizing NANOSENSORS™ PPP-NCHR AFM probes. If necessary, the AFM images were processed by flattening to remove the background slope and adjusted for contrast and brightness.
# Results and discussion
We seek to create a supramolecular structure in which one oligonucleotide contributes a run of C's to an i-motif and a run of G's to a G4. We use mass spectrometry to observe multimeric complexes of the oligonucleotide. I-motif anions will have a mass that is simply the sum of the four strands with the appropriate number of hydrogens to give the observed charge. G4 anions require n-1 cations (here NH 4 + ) for n G-quartets assembled. The structural arrangement is confirmed by CD. Since i-motifs only form under acid conditions, the CD difference spectrum resulting from a pH 9.0 spectrum subtracted from a pH 4.5 spectrum reveals the imotif contribution. For G4s, we use cation dependence: K + supports parallel quartet stacking, while Na + shows preference for antiparallel stacking and NH 4 + falls in between these two cations based on the nature of sequences, but Li + does not support any stacking of G-quartets [bib_ref] Guanine quartet structures, Sen [/bib_ref] [bib_ref] Tetramolecular quadruplex stability and assembly, Tran [/bib_ref]. Thus, the CD difference spectrum of a sample in the presence of NH 4 + minus the spectrum in Li + , gives the G4 contribution. See SI for a table of characteristic CD peaks for different DNA structures.
## The influence of the location of guanine residues or gstretches on the tetramolecular i-motif formation
Suitable solution conditions for the formation of tetrameric i-motifs were determined by studying the oligonucleotide d(TC [bib_ref] Human telomeric DNA: G-quadruplex, i-motif and Watson-Crick double helix, Phan [/bib_ref] We investigated the influence of guanine residues or Gruns on the formation of i-motif structures and the competition between WC duplex and i-motif structures using the sequences d(TG i TC 5 T) and d(TC 5 TG i T) (i = 1, 2, 3). The spectra in Supplementary [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref] all indicate the presence of i-motif tetramers [bib_ref] Circular dichroism and conformational polymorphism of DNA, Kypr [/bib_ref] and that they are more prominent in d(TGTC 5 T). We have previously demonstrated that short C-rich strands facilely form duplexes with C- CH + pairs under acidic condition and can be observed along with tetrameric i-motifs [bib_ref] Formation and dissociation of the interstrand i-motif by the sequences d(X n..., Cao [/bib_ref]. Consequently, the low-abundance duplex ions observed here are mainly C- CH + -containing duplexes (C-duplex) since there are insufficient complementary G and C bases between two oligonucleotides to create WC duplexes (Supplementary [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref].
The two oligonucleotides d(TG 2 TC 5 T) and d(TC 5 TG 2 T) show the same tendency, with an even greater inhibition of i-motif formation by a 3 tailing G 2 T. The d(TG 2 TC 5 T) sequence assembles into a tetrameric i-motif structure in acidic solution, as supported by the observation of . We propose that the bimolecular ions observed under acidic conditions may correspond to two types of duplex structures: a C- CH +containing-duplex as well as a bulged WC duplex (the Cduplex with five non-intercalated C- CH + base pairs will lose another five hydrogen cations to give the same m/z as the bulged WC duplex). In contrast, the bimolecular ions formed under alkaline conditions mainly correspond to the bulged WC duplex, which has no hemi-protonation of cytosines (Supplementary [fig_ref] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs [/fig_ref]. The percentage of [TC 5 TG 2 T] 2 ions decreased from 42.56 to 9.99%, when the solution pH was adjusted from 4.5 to 9.0 [fig_ref] Table 2: The percent abundance of various ion species formed at different pH's as... [/fig_ref]. If it is assumed that the same fraction of the samples ionize, we can conclude that the proportion of C-duplex and bulge duplex is 3.3 : 1 (32.57% : 9.99%) under acidic condition when the influence of solution pH on the formation of WC duplexes is ignored. Since CD signal of the non-intercalative C- CH + base pair stacking may be weak (48), the CD spectrum of d(TC 5 TG 2 T) only displays a single positive band around 290 nm. However, the CD difference spectrum obtained by subtracting the pH 9.0 spectrum from the pH 4.5 spectrum , inset) shows the positive band at 288 nm and the negative band at 265 nm, suggestive of parallel C-duplex. In contrast, i-motif structures were not formed effectively in acidic solutions when the G-stretches extended to three guanines. : 1) for d(TG 3 TC 5 T) and 12.89% : 10.5% (≈ 1.2 : 1) for d(TC 5 TG 3 T) under acidic conditions. The CD difference spectrum of d(TG 3 TC 5 T) , inset) displays the characteristic CD signature of protonated C-duplex with an approximate 5 nm blue shift for the negative band at 265 nm, while d(TC 5 TG 3 T) shows the two typical peaks of protonated C-duplex, each blue shifted by approximately 5 nm in the CD difference spectrum , inset). Presumably the increase in complementarity allows the formation of more bulged WC duplexes. In addition, CD spectrum of d(TG 3 TC 5 T) shows that a small negative band around 210 nm coexists with the positive band around 260 nm , which implies that the oligonucleotide of d(TG 3 TC 5 T) not only forms the B-form double helix but also forms the A-form hairpin structure (Scheme 2D) as has been described by others [bib_ref] Hairpin-duplex equilibrium reflected in the A→B transition in an undecamer quasi-palindrome present..., Kaushik [/bib_ref]. In summary, we conclude that runs of three G's are detrimental to i-motif formation and that the introduction of G-repeats at the 3 end hinders the formation of i-DNA tetramers more significantly than their introduction at the 5 end.
We assume that the i-motif structure formed by d(TG 2 TC 5 T) sequence will have two sets consecutive G's at each terminus (Scheme 2A) available for interchain interactions. The formation of a parallel G4 goes through dimer and trimer intermediates [bib_ref] The formation pathway of tetramolecular G-quadruplexes, Bardin [/bib_ref] [bib_ref] Tetramolecular G-quadruplex formation pathways studied by electrospray mass spectrometry, Rosu [/bib_ref]. However, two parallel dangling 5 -GGT's are unlikely to pair with one another since the tetrameric G4 with only two G-quartets is known to be very unstable [bib_ref] Pyrene excimer fluorescence as a probe for parallel G-quadruplex formation, Zhu [/bib_ref].
## The effect of central and terminal thymines on tetramer formation
The oligonucleotide d(TG 3 TC 5 T) has little tendency to form i-motif structure in our experimental conditions. We investigated whether or not the spacer nucleotide at the junction between G-and C-repeats influenced i-motif formation. First we studied the strand d(TG 3 C 5 T), which lacks a spacer base. Supplementary [fig_ref] Figure 4: Product-ion spectra of d [/fig_ref] shows that the [fig_ref] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs [/fig_ref] and Data of [fig_ref] Table 2: The percent abundance of various ion species formed at different pH's as... [/fig_ref]. These results demonstrate that the WC duplex formation also becomes more favorable by removing the spacer base. This is likely due to eliminating the mismatch in the middle of doublehelix structures.
We then tried d(TG 3 AC 5 T), which substitutes the spacer thymine with adenine, the only other base that does not extend either the G-or C-run. [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref] (26), while the characteristic peaks of A-type-duplex-containing hairpins are not observed [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref]. We propose that d(TG 3 AC 5 T) forms tetrameric i-motifs because the adenine purine supports stacking over bending into the loop of a single-strand hairpin structure [bib_ref] Influence of loop residues on the relative stabilities of DNA hairpin structures, Senior [/bib_ref]. Furthermore, the steric hindrance of paired adenines may be unfavorable to the formation of a bulged WC duplex.
It has been known that the flanking residues of DNA tetramers at both ends have significant effect on the strand association and structural stability [bib_ref] Acid multimers of oligodeoxycytidine strands: stoichiometry, base-pair characterization, and proton exchange properties, Leroy [/bib_ref]. Hence the 3 end thymine of d(TG 3 AC 5 T) was removed. [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref] and D show that the tetramolecular ions of d(TG 3 AC 5 ) are more abundant than that of d(TG 3 AC 5 T) (the percentage of tetramolecular ions changes from 39.22 to 79.95%). The CD spectra in [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref] indicate a sharp enhancement of the i-motif peaks under acidic conditions when the 3 thymine is removed. In comparison, the removal of the 5 thymine has little effect on the i-motif formation: both the abundance of tetramolecular ions in [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref] (the percentage of tetramolecular ions is 29.11%) and the CD signals of i-motif structure in [fig_ref] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences [/fig_ref] have a slight decrease compared with the sequence d(TG [bib_ref] G-quadruplex DNA assemblies: loop length, cation identity, and multimer formation, Smargiasso [/bib_ref]
## Ac 5 t).
This indicates that d(TG 3 AC 5 ) is most amenable to tetrameric i-motifs formation. I-motif structures can be classified into two types, 3 E and 5 E, determined by the outside terminus of the intercalated C- CH + base pairs [bib_ref] Investigation of the energetics of C-H···O hydrogen bonds in the DNA i-motif..., Leroy [/bib_ref]. 3 E and 5 E i-motif structures with or without 3 thymines of d(TG 3 AC 5 T) are depicted in Scheme 3. In both 3 E and 5 E, the C- CH + base pairs from two orthogonal C-duplexes stack to place the 3 thymine residues far from the central adenine residues. The distance between them is about 0.31 nm, which is similar to the space between two WC base pairs (about 0.34 nm) [bib_ref] Investigation of the energetics of C-H···O hydrogen bonds in the DNA i-motif..., Leroy [/bib_ref] [bib_ref] Molecular structure of nucleic acids, Watson [/bib_ref]. Consequently, the 3 thymine and central adenine residues do not easily form A- T WC base pairs to further stabilize the i-motif structure. Furthermore, the 3 thymine residues crowd the terminal regions of i-motif structures. A recent kinetics investigation demonstrated that the association rates of tetrameric i-motifs with partial (out of register) or full intercalation topology are comparable [bib_ref] The formation pathway of i-motif tetramers, Leroy [/bib_ref]. By reducing the 3 thymine residues, various i-motifs with partial registry will form along with the fully intercalated (registered) species without restrictions. This dynamically favors the formation of i-motifs when they compete with WC duplex structures. For these reasons, removing the 3 thymine can promote the formation of i-motif structure dramatically.
## Controlled assembly of tg m ac n quadruplex dna nanoarchitectures
In our experiments, d(TG 3 AC 5 ) forms a significant amount of i-motif with two parallel G-runs at each end. These parallel G-runs should serve as flexible branches to link two i-motif scaffolds through the formation of interstrand antiparallel or parallel G4s, as depicted in Scheme 4. Antiparallel association should give DNA quadruplex nanowires with alternating G4s and i-motifs, while parallel association would create G-quartet hubs surrounded i-motifs. Since both i-motifs and G4s have slow assembly kinetics [bib_ref] The formation pathway of i-motif tetramers, Leroy [/bib_ref] [bib_ref] Kinetics of tetramolecular quadruplexes, Mergny [/bib_ref] , it is possible that single i-motifs and G4s might be observed simultaneously as the oligonucleotide assembles into a supramolecular structure. These may be independent assemblies or partially associated supramolecular complexes that break down during ionization.
No tetrameric ions with the appropriate number of cations corresponding to G4s were observed in mass spectra of d(TG 3 AC 5 ). It could be that the three-nucleotide G-run is too short to support a stable tetrameric G4 in this context. Hence, we extended the length of G-and Cruns to d(TG 4 AC 6 ). Supplementary [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref] shows that under acidic conditions allowing i-motifs, this strand produced tetrameric ions with three NH [bib_ref] Stability and molecular recognition of quadruplexes with different loop length in the..., Arora [/bib_ref] [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref]. Unfortunately, these tetramer species only represent a small proportion of the sum of all strands (17.57%). In particular, there are still strong peaks from monomers and dimers, indicating that a significant part of the sample is not in a tetraplex or more complex assemblage. Supplementary [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref] shows that NH 4 + can enhance the CD signatures at 220, 245 and 265 nm obviously in comparison with Li + , suggesting the formation of parallel G4 in NH 4 + solution. The difference spectrum of Li + subtracted from NH 4 + containing solutions, both taken at pH 4.5, does not give a typical profile of a parallel G4, which may be due to less abundant G4s formed (Supplementary [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref].
We added a central adenine to give d(TG 4 A 2 C 6 ). Supplementary [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref] show that no tetrameric ions are observed for this strand in the presence NH 4 + under both pH conditions. Solutions of this strand also produced no change in the CD signal when Li + was substituted for NH 4 + (Supplementary [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref]. The additional adenine is detrimental to tetramer formation. The oligonucleotide with only one spacer A and one more C, d(TG 4 AC 7 ), produces an abundance of tetrameric ions [fig_ref] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs [/fig_ref] [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref]. NH 4 + -adductfree tetrameric i-motif ions were not observed under this condition; however, the CD difference spectrum obtained by subtracting the pH 9.0 spectrum from the pH 4.5 spectrum indicates the presence of C- CH + base pairs (Supplementary [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] , inset). Under acid conditions decreasing NH 4 + ions from 50 to 30 mM, the lower abundant ions corresponding to i-motif tetrameric ions were observed (the ions with m/z at 2596.71 for 6-and at 2225.60 for 7-) along with G4 tetrameric ions; continuously decreasing NH 4 + ion concentrations, at 20 mM NH 4 + , the abundance of tetramolecular ions containing three NH 4 + ions decreased drastically; at 10 mM NH 4 + , the G4s were even more unstable and only bimolecular ions were observed (data no shown). The G4 monomer units of d(TG 4 AC 7 ) are more easily observed than i-motif monomer units, which may be due to a number of factors. G4s with protonated C-duplexes may be more stable than i-motifs with G-rich overhangs (Scheme 4A versus B), perhaps because the C's sequestered in C- CH + pairs, and were unable to form WC pairs with G's [bib_ref] Cation-dependent transition between the quadruplex and Watson-Crick hairpin forms of d(CGCG 3..., Hardin [/bib_ref] [bib_ref] Cytosine-cytosine + base pairing stabilizes DNA quadruplexes and cytosine methylation greatly enhances..., Hardin [/bib_ref]. When alkaline conditions release these C's they readily form WC pairs with the G's, preventing quadruplex formation.
Tetramolecular ions adducted with NH 4 + observed in our mass spectrum show good agreement with the G4s formed by d(TG 4 AC 7 ) that specifically have three buried Scheme 4. Postulated quadruplex structures for d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ): (A) the tetrameric i-motif with G-rich overhangs at each terminal; (B) tetra-stranded parallel and antiparallel G-quadruplexes (G4s) with two pairs of protonated C-duplexes formed by d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ) respectively under acidic conditions; circle: purine; triangle: pyrimidine; in a G-tetrad, anti guanosines are indicated in white and syn guanosines are indicated in black; Supramolecular DNA structure that could assemble from oligonucleotide with sequences (C) d(TG 4 AC 7 ) constructed by parallel tetrastrand G4s and antiparallel i-motifs; and (D) d(TG Br GG Br GAC 7 ) formed by antiparallel tetra-strand G4s, antiparallel i-motifs and (or) parallel G4s. NH 4 + ions among four stacked G-quartets [fig_ref] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs [/fig_ref]. In order to show that these NH 4 + ions are not the ions adsorbed via electrostatic interactions with the phosphate backbone, MS/MS experiments for the d(TG 4 AC 7 ) [4M + 3NH 4 ] xions were performed. [fig_ref] Figure 4: Product-ion spectra of d [/fig_ref] shows production spectra of the d(TG 4 AC 7 ) [4M + 3NH 4 ] 7ion (m/z = 2232.77) resulting from collision energies of 10, 20 and 30 eV. The separation of NH 4 + ions is not significant until the collision energy reaches 30 eV. This energy is sufficient to disrupt interchain interactions [bib_ref] Formation and dissociation of the interstrand i-motif by the sequences d(X n..., Cao [/bib_ref] ; hence, the NH 4 + ions can only dissociate upon disruption of the tetraplex.
I-motif formation and stability were investigated with UV melting experiments at 265 and 295 nm for the sequence d(TG 4 AC 7 ) in pH 4.5 LiOAc buffer. Both melting curves give two apparent melting points around 50 - C and 75 - C, which are considered to be the dissociations of duplexes (including WC duplex and C-duplex) and i-motif structures, respectively (Supplementary . This conclusion is further supported by the UV melting experiment at pH 9.0 in LiOAc buffer solution, . It shows only a single transition point at 260 and 265 nm, corresponding to the dissociation of WC duplex and the melting temperature (T m ) of WC duplex obtained from the UV melting curve (at 260 and 265 nm) is about 60 - C, while no obvious transition point is observed for the measurement at 295 nm. It is unwise to further increase the length of C-repeats since this may induce the formation of an intramolecular C + - GC triplex (58). We next investigated the effect of modification of guanines on the formation and stability of tetramolecular G4s. It has been reported that the inclusion of two 8methyl-2 -deoxyguanosine (M) residues in d(TMGMGT) promotes the formation of an unprecedented antiparallel tetra-stranded G4 in which two adjacent strands run in the same direction [bib_ref] The insertion of two 8-methyl-2 -deoxyguanosine residues in tetramolecular quadruplex structures: trying..., Virgilio [/bib_ref]. Since a bromine atom has a similar size as a methyl group and the 8-bromoguanine residue has the same tendency to promote a syn glycosidic conformation [bib_ref] Carbon-13 magnetic resonance spectra of 8-substituted purine nucleosides. Characteristic shifts for the..., Uesugi [/bib_ref] [bib_ref] Effects of an 8-bromodeoxyguanosine incorporation on the parallel quadruplex structure, Esposito [/bib_ref] [bib_ref] 8-methyl-2 -deoxyguanosine incorporation into parallel DNA quadruplex structures, Virgilio [/bib_ref] , we investigated the assembly properties of d(TG Br GG Br GAC 7 ), where G Br indicates an 8-bromoguanine residue. [fig_ref] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs [/fig_ref] [fig_ref] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs [/fig_ref] , inset) suggests that the tetrameric G4 is antiparallel as depicted in Scheme 4B based on previous reports [bib_ref] The insertion of two 8-methyl-2 -deoxyguanosine residues in tetramolecular quadruplex structures: trying..., Virgilio [/bib_ref] [bib_ref] Development of new functional nanostructures consisting of both DNA duplex and quadruplex, Dutta [/bib_ref]. As with d(TG 4 AC 7 ), the tetramers were dissociated under alkaline conditions (Supplementary [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] and the CD difference spectrum obtained by subtracting the pH 9.0 spectrum from the pH 4.5 spectrum indicated the presence [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] , inset). We are not able to observe larger species formed by d(TG 4 AC 7 ) or d(TG Br GG Br GAC 7 ) due to our instrument's m/z range limit.
The formation of quadruplex DNA nanoarchitectures was further indicated by non-denaturing polyacrylamide gel electrophoresis (PAGE). K + ions can stabilize parallel G4s effectively, while Na + and K + ions have distinct effects on various types of antiparallel G4s (3). Consequently, d(TG 4 AC 7 ) was mainly annealed in KOAc buffer solutions at pH 4.5 and pH 9.0 to ensure the desirable stability of the parallel G4 formed by GGGG-fragment in d(TG 4 AC 7 ), and d(TG Br GG Br GAC 7 ) was annealed in both KOAc and NaOAc buffer at the two pH's for the PAGE experiment since both parallel and antiparallel G4s may be formed by G Br GG Br G-fragment in d(TG Br GG Br GAC 7 ). These gels run at pH 4.5 and 4 - C in order to stabilize i-motifs. We believe that samples from higher pH's form only a small amount or no supramolecular DNA structures due to the pH drop upon loading and running in the gel, because the rate of formation of such supramolecular DNA structures is very slow at low temperature [bib_ref] association into supra molecular i-motif structures, Laisné [/bib_ref].
In accordance with the UV melting study, [fig_ref] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4 [/fig_ref] shows that d(TG 4 AC 7 ) incubated in pH 4.5 LiOAc solution has three obvious bands corresponding to single-, doublestranded structures and tetra-stranded i-motifs. However, d(TG 4 AC 7 ) incubated in pH 9.0 KOAc solution has only two obvious bands corresponding to single-and doublestranded structures. The various smeared bands migrating slowly are believed to be supramolecular structures assembled by the oligonucleotide at pH 4.5 in the presence of K + . Similar results were also obtained for d(TG Br GG Br GAC 7 ). The bands of single-stranded structures are much broader than the double-stranded structures under alkaline conditions. They are likely to contain two components: the open random coil conformation and the hairpin conformation, which may be exchanging. d(TG Br GG Br GAC 7 ) forms larger structures at pH 4.5 with either Na + or K + ions. When K + ions are present the structures stay at the top of the gel. Under this condition both parallel and antiparallel G4s can form. The dendritic clusters from parallel G-quartet assembly (Scheme 4D, right) may not enter the gel. It may also be that the supramolecular structures formed at pH 4.5 with K + are more stable.
Lastly, we provide evidence for the formation of DNA supramolecular structures directly using AFM. [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] shows several small prolate ellipsoids [blobs] and V-shaped and ring-like aggregates formed at pH 4.5 in KOAc buffer by d(TG 4 AC 7 ). The ellipsoids have diameters (height) around 3.2 nm and lengths of up to 60 nm. These are consistent with the dimensions of individual tetra-strand G4s or imotifs, while the V-shaped and ring-like aggregates (4.5 nm in height) are consistent with larger multimers of tetrastrand G4s and i-motifs [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] such as those depicted in Scheme 4C [bib_ref] The I-tetraplex building block: rational design and controlled fabrication of robust 1D..., Ghodke [/bib_ref] [bib_ref] Self-assembled G-quadruplex nanostructures: AFM and voltammetric characterization, Chiorcea-Paquim [/bib_ref]. As expected, these structures were not observed at pH 9.0, suggesting the dissociation of these high-order structures and possible inhibition of tetrameric structures by competing WC interactions within and between strands [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref]. We propose that the nanostructures formed by d(TG 4 AC 7 ) are mainly constructed by parallel G4s and antiparallel i-motifs. In this case, the tetramers will assemble to be a 'V-shaped' structure in first step of polymerization. Meanwhile, the 'V-shaped' structures will further assemble to be a ring-like texture.
When the counter ion is Na + , the structures formed at pH 4.5 are less regular and only some small ellipsoids and V-shaped aggregates, which may be catenated, were ob-served (Supplementary Data, Supplementary . This was in agreement with the reports that the K + stabilizes parallel G4s better than Na + [bib_ref] Selective binding of monovalent cations to the stacking G-quartet structure formed by..., Wong [/bib_ref].
Even larger aggregates are seen by AFM with d(TG Br GG Br GAC 7 ) [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] , including V-shaped and rod-like shape aggregates with an average height of 4.0 nm. These may be quadruplex nanowires [bib_ref] The I-tetraplex building block: rational design and controlled fabrication of robust 1D..., Ghodke [/bib_ref] [bib_ref] Self-organization of guanosine 5 -monophosphate on mica, Kunstelj [/bib_ref] [bib_ref] Programmed self-assembly of a quadruplex DNA nanowire, Hessari [/bib_ref] with alternating G4s and i-motifs. [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref] indicates that the lengths of these rod-like shape aggregates can get as long as 550 nm. Just as before, the quadruplex nanowires dissociate under alkaline conditions [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref]. Both the parallel C-duplexes and the two G-repeats at each end are excellent junctions to further propagate the linear supramolecular DNA structures [fig_ref] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex... [/fig_ref]. It should be noted that the V-shaped aggregates are also observed for d(TG Br GG Br GAC 7 ) in pH 4.5 KOAc solution, indicating that both parallel and antiparallel four-stranded G4s form in the presence of K + (Scheme 4D, right). However, only some small prolate ellipsoids and rod-like aggregates were observed in AFM image when the d(TG Br GG Br GAC 7 ) strand was annealed in pH 4.5 NaOAc buffer (Supplementary .
We also investigated the solution-state behavior of three sequences d(TC 5 T), d(TG 3 AC 5 ) and d(TG 4 AC 6 ) observed by AFM. -c show that only small prolate ellipsoids are formed by d(TC 5 T) and small amounts of V-shaped aggregates are produced by d(TG 3 AC 5 ) and d(TG 4 AC 6 ) respectively in pH 4.5 KOAc solutions, indicating that these sequences are unfavorable to form larger aggregates. Additionally, we investigated the effect of cation species and pH on the control of assembly of the supramolecular structures formed by d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ). S9d and g show that some prolate ellipsoids formed by d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ) respectively at pH 4.5 in LiOAc buffer solutions correspond to tetra-stranded i-motifs. These prolate ellipsoids can further assemble to give V-shaped or rod-like supramolecules when K + ions are added into the pH 4.5 LiOAc solutions as shown in Supplementary and h. Similar results are also obtained when d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ) are annealed in pH 9.0 KOAc solution, and then the solution pH is adjusted to 4.5 (Supplementary . Hence, we propose that the assembly of quadruplex supramolecules can be controlled or reproduced; however, the structural transitions are less efficient in comparison with the assembling of them under optimum conditions, which may indicate a slow shift process.
# Conclusions
A sequence with a long C-run and a shorter G-run can simultaneously form various DNA structures, including a hairpin, a WC duplex, a G4 and an i-motif under different conditions. This polymorphic property gives such oligonucleotides the potentiality to form supramolecular nanostructures. We found sequences that support the stable formation of tetra-stranded i-motifs with G-repeats at two ends by altering the position of G-repeats, by changing the linker between G-and C-repeats, and removing the terminal base of the C-repeat. Ultimately, we find that the sequences d(TG m AC n ) (m < n) have the ability to form four-stranded i-motif structures with relative long G-repeats at each terminus. Two sequences d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ) were investigated under various conditions, utilizing the ability of each G-repeat to form stable parallel and antiparallel tetramolecular G4s, respectively. What is noteworthy is that both of two sequences form quadruplex nanowire structures under acidic conditions in the presence of specific metal cations. The nanowire structures formed by d(TG 4 AC 7 ) are composed of antiparallel tetrameric imotifs and parallel tetrameric G4s, which take on a Vshaped conformation (most likely two i-motif tetra-strand units and one G4 tetra-strand unit; Scheme 4C middle) and ring-like conformations (including additional i-motif and G4 tetra-strand units; Scheme 4C right). These aggregates resemble the products of DNA origami. On the other hand, the nanowire structures formed by d(TG Br GG Br GAC 7 ) are made up of antiparallel tetra-strand i-motifs and antiparallel G4s to give linear DNA pillars stabilized by H + and Na + ions. Studies have shown that a single G-strand consisting of a number of 4n G-repeats can form higher-ordered structures containing n G4s [bib_ref] Single molecule studies of physiologically relevant telomeric tails reveal POT1 mechanism for..., Wang [/bib_ref] [bib_ref] Structure and stability of higher-order human telomeric quadruplexes, Petraccone [/bib_ref] [bib_ref] Tetrahelical monomolecular architecture of DNA: a new building block for nanotechnology, Kankia [/bib_ref]. These require the synthesis of very long DNA pieces. Here, our results demonstrated that four-stranded G4s or (and) i-motifs formed from relatively short (G + C)-rich oligonucleotide sequence could also self-assemble to drive supramolecular structures under specific conditions. Furthermore, the association and dissociation of these nanostructures can be modulated by solution pH and metal cations, which may provide a mechanism for controllable molecular machines.
[fig] 4, Figure 1: and D show that only bimolecular ions (including [2M] 5− at m/z 1312.45, [2M] 4− at m/z 1640.81 and [2M] 3− at m/z 2188.35) were detected in mass spectra both for the sequences d(TG 3 TC 5 T) and d(TC 5 TG 3 T). Scheme 2B and C propose two possible structures for dimers of d(TG 3 TC 5 T): a parallel C-duplex or a bulged double helix. Data of Table 2 indicates that the proportions of these two structures are 20.39% : 15.04% (≈ 1.Mass spectra of assemblies of oligonucleotides with sequences (A) d(TG 2 TC 5 T), (B) d(TC 5 TG 2 T), (C) d(TG 3 TC 5 T) and (D) d(TC 5 TG 3 T) at 60 M in 50 mM NH 4 OAc (pH 4.5) after incubating the samples at 4 • C for 4 days; (E-H) CD spectra of the four oligonucleotides in the acidic (solid line) and alkaline solutions (dotted line); inset: subtracted CD spectrum obtained by subtracting the CD spectrum in alkaline buffer solution from the CD spectrum in acidic buffer solution. Scheme 2. Schematic drawings of (A) 3 E (end) fully intercalated tetramolecular i-motif formed by d(TG 2 TC 5 T); (B) C•CH + base pair duplex; (C) and (D) WC double helix and hairpin structure with different lengths of flanking residues formed by d(TG 3 TC 5 T); solid wedged bond represents intercalation C•CH + base pair; dashed wedged bond represents non-intercalation; solid stick bond represents WC pairing. [/fig]
[fig] Figure 2: Mass spectra of assemblies of oligonucleotides with sequences (A) d(TG 3 AC 5 T), (B) d(TG 3 AC 5 ) and (C) d(G 3 AC 5 T) at 60 M in 50 mM NH 4 OAc (pH 4.5) after incubating the samples at 4 • C for 4 days; (D) CD spectra of the three DNAs at pH 4.5. [/fig]
[fig] Scheme 3: Schematic drawings of 3 E and 5 E fully intercalated tetramolecular i-motifs (A) with 3 thymines [d(TG 3 AC 5 T)] and (B) without 3 thymines [d(TG 3 AC 5 )]. [/fig]
[fig] Figure 3: Mass spectra of assemblies of oligonucleotides with sequences (A) d(TG 4 AC 7 ) and (B) d(TG Br GG Br GAC 7 ) at 60 M in 50 mM NH 4 OAc (pH 4.5) after incubating each sample at 4 • C for 4 days; (C and D) CD spectra of the two sequences in pH 4.5 NH 4 OAc and pH 4.5 LiOAc buffers; inset: CD difference spectrum obtained by subtracting the CD spectrum with the LiOAc solution from the CD spectrum with the NH 4 OAc solution. [/fig]
[fig] Figure 4: Product-ion spectra of d(TG 4 AC 7 ) [4M+3NH 4 ] 7− ion (m/z = 2232.77) for collision energies of (A) 10 eV, (B) 20 eV and (C) 30 eV. [/fig]
[fig] Figure 5: Native electrophoretic mobility on 8% polyacrylamide at pH 4.5 of oligonucleotides d(TG 4 AC 7 ) and d(TG Br GG Br GAC 7 ) in different buffers containing 100 mM cations after annealing DNA samples at 4 • C for 1 week: Lane 1: a molecular mass ladder: 13, 26 and 39 bp; Lane 2: d(TG 4 AC 7 ) with pH 4.5 LiOAc; Lane 3: d(TG 4 AC 7 ) with pH 9.0 KOAc; Lane 4: d(TG 4 AC 7 ) with pH 4.5 KOAc; Lane 5: d(TG Br GG Br GAC 7 ) with pH 9.0 KOAc; Lane 6: d(TG Br GG Br GAC 7 ) with pH 4.5 NaOAc; Lane 7: d(TG Br GG Br GAC 7 ) with pH 4.5 KOAc; ss, single-stranded structures; sms, supramolecular structures. of C•CH + base pairs under acidic conditions (Supplementary [/fig]
[fig] Figure 6: AFM images with the scale bar of 200 nm for DNA quadruplex wires self-assembled from oligonucleotide sequences d(TG 4 AC 7 ) (A and C) and d(TG Br GG Br GAC 7 ) (D and F) with different solutions and height scales: (A and D) pH 4.5 KOAc buffer; (C and F) pH 9.0 NaOAc solution. The height and length profiles of the aggregates formed by d(TG 4 AC 7 ) (B) and d(TG Br GG Br GAC 7 ) (E) deposited on the freshly cleaved mica substrate follows the lines on (A) and (C) respectively. The triangles indicate the same points in the image and the height analysis. The arrows in (A) are used to point to DNA structures with different shapes and sizes. [/fig]
[table] Table 1: Sequences of oligodeoxynucleotides studied here [/table]
[table] Table 2: The percent abundance of various ion species formed at different pH's as measured by mass spectrometry for the various oligonucleotides studied Each data obtained is the average of two measurements.tetrameric ions ([4M] 7− at m/z 1687.57 and [4M] 6− at m/z 1968.99) and the characteristic CD signature of the i-motif structure (Figure 1A and E). However, d(TC 5 TG 2 T) only produces the duplex ions ([2M] 5− at m/z 1181.01, [2M] 4− at m/z 1476.26 and [2M] 3− at m/z 1968.68) and no i-motif tetrameric ions were observed ( [/table]
|
Abscess formation mimicking disease progression, in a patient with metastatic renal cell carcinoma during sunitinib treatment
Background: Renal cell carcinoma (RCC) represents approximately 3% of all adult cancers and is more common in males. Systemic treatment for RCC has improved following the introduction of tyrosine kinase inhibitors, such as sunitinib. The molecular targets of sunitinib are receptor tyrosine kinases (RTKs). Moreover, sunitinib has an additional anti-angiogenic effect through its inhibition of the vascular endothelial growth factor receptor activation.Case presentation:We present a case of intra-abdominal abscess formation mimicking disease progression, in a patient with metastatic renal cell carcinoma during sunitinib treatment.Conclusion:In the advancing era of molecular therapy of solid tumours, sunitinib has demonstrated significant efficacy in the post-cytokine setting treatment of metastatic renal cancer. Concurrently, however, increasing evidence has emerged to indicate that this class of drugs exert profound immunomodulatory effects on T cells and play major roles in immune tumor surveillance.
# Background
The treatment of advanced RCC is undergoing a paradigm shift with the recent introduction of anti-angiogenic therapy that either directly inhibits vascular endothelial growth factor or disrupts signal transduction favorable to vascular development through multi-kinase inhibitors. Angiogenic inhibitors have been found to increase survival and are approved in advanced renal cell carcinoma [bib_ref] TARGET Study Group: Sorafenib in advanced clear-cell renal-cell carcinoma, Escudier [/bib_ref] [bib_ref] Sunitinib versus interferon alfa in metastatic renal-cell carcinoma, Motzer [/bib_ref]. Consequently, most of these patients will routinely receive tyrosine kinase inhibitors, such as sunitinib.
Sunitinib is an orally administered, small molecule inhibitor of multiple receptor tyrosine kinases implicated in tumour growth, angiogenesis, and metastatic progression. In addition, the targets of sunitinib involve vascular endothelial growth factor receptors (VEGFR1, VEGFR2 and VEGFR3), platelet-derived growth factor receptors (PDGFRα and PDGFRβ) and the like. We describe a case of intra-abdominal abscess formation mimicking disease progression during sunitinib treatment.
## Case presentation
A 62-year-old patient diagnosed with a high-grade clearcell renal carcinoma in 1991 and was treated by left nephrectomy and surrenalectomy. Fourteen years later, relapsed on the lungs and had been administered interferon alfa. The patient was regularly followed up and had regular scans that did show stabilization of the disease in the lungs for two years. In December 2007 chest computerized tomography (CT) disclosed the progression of lung metastases. Sunitinib was initiated in January 2008 as a standard regimen (50 mg/day for 4 weeks every 6 weeks) for pulmonary metastases. Patient had a radiographic response and prolonged progression free survival of fourteen months; side effects were manageable and included grade 2 hypertension. After five cycles, the patient was admitted to the hospital due to complaints of fatigue and left sided flank pain. The systolic and diastolic blood pressures were 110 mmHg and 60 mmHg, respectively, pulse rate was 90 per min and respiratory rate was 20 per min. The body temperature was 37.2°C.
Laboratory studies were conducted immediately after the patient's arrival at the hospital. He had anemia (Hb 98 g/L) (normal range: 140-180) and thrombocytopenia (133 × 10 9 /L) (normal range: 150-450), but a WBC count was normal (6.15 × 103/mm 3 ) with 74% neutrophils. Other laboratory findings were presented as elevated serum levels of CRP (21 mg/L) (normal range:< 5), ALP (416 IU/L) (normal range: 96-250), and slightly increased creatinine (1.43 μmol/L) (normal range: 0.5-1.2).
Fluorodeoxyglucose positron emission tomography (FDG-PET-CT) scans demonstrated an area of increased uptake in the left paravertebral area [fig_ref] Figure 1: PET -CT demonstrating an area of increased uptake in the left paravertebral... [/fig_ref]. MRI scan (T2 image) demonstrated the lesion that corresponded to the area of increased PET uptake [fig_ref] Figure 2: MRI scan [/fig_ref].
The patient underwent a diagnostic laparoscopy in May 2009. Intra-operative biopsy of the lesion was performed; the pathology was consistent with an abcess without evidence of malignancy. After an uneventful postoperative course, the patient was discharged on the 10th day after surgery and chemotherapy with sunitinib was restarted. Three months postoperatively there was no evidence of recurrent disease.
# Discussion
Renal cell cancer (RCC) is a relatively uncommon malignancy. When the disease is localized is curable by surgery; however, locally advanced or metastatic disease is not curable in most cases and until recently had a limited response to drug treatment. Historically, biologic response modifiers or immunomodulating agents were tested in clinical trials based on observations that some cases of RCC can spontaneously regress. Responses have been observed with interferon alfa, but with little effect on overall survival.
The use of targeted therapies has substantially improved outcomes for patients with advanced renal cell carcinoma [bib_ref] Role of Raf kinase in cancer: therapeutic potential of targeting the Raf/MEK/ERK..., Gollob [/bib_ref] [bib_ref] Long-term Response with Sunitinib for Metastatic Renal Cell Carcinoma, Ronnen [/bib_ref].
Sunitinib malate is an oral multi-kinase inhibitor targeting several receptor tyrosine kinases (PDGFRalpha and PDGFRbeta; VEGFR1, VEGFR2 and VEGFR3; KIT, FLT3, CSF-1R and RET) that was approved by the FDA in 2006 for treatment of metastatic renal cell carcinoma. In a randomized phase III trial, sunitinib prolonged median progression-free survival (11 months) in comparison to interferon-alpha (5 months); corresponding to a hazard ratio of 0.42 (95% confidence interval: 0.32 to 0.54; P < 0.001) for patients with advanced renal cell cancer. Sunitinib was also associated with a higher objective response rate than interferon-alpha (31% vs. 6%; P < 0.001) [bib_ref] Sunitinib versus interferon alfa in metastatic renal-cell carcinoma, Motzer [/bib_ref].
The most common toxicities with sunitinib are handfoot syndrome, rash, fatigue, hypertension, and diarrhea.
Concurrently, however, increasing evidence has emerged to indicate that TKIs such as sunitinib, exert profound immunomodulatory effects on T cells and antigen-presenting cells, such as dendritic cells, which play major roles in immune tumor surveillance. Targeted tyrosine kinase inhibitor therapy may thus control cancer cell growth both directly and indirectly by changing the immunologic microenvironment. These side-effects of therapy on normal vasculature type may lead to rare complications such as the abscess formation in the present case. However, physicians should have in mind that occasionally disease extension to unexpected anatomical sites does occur, causing unusual clinical pictures. For the differential diagnosis between these conditions, CT scan is considered to be the imaging study with the highest accuracy and efficiency [bib_ref] Renal and perirenal abscesses, Dembry [/bib_ref] [bib_ref] Renal abscess, Geeting [/bib_ref]. Not only can it be of great help in diagnosis, but also in evaluating the extension of involvement. Furthermore, an approach for drainage of abscesses can be made on CT results. However, sometimes an exploratory laparotomy is necessary to reveal the cause. Nonetheless, in our case, CT findings were not sufficient for the diagnosis and the cause of imaging findings was unclear until laparotomy.
In summary, a case of an intra-abdominal abscess formation mimicking disease progression during sunitinib treatment, was presented. After an uneventful postoperative course, the patient was discharged on the 10th day after surgery and chemotherapy with sunitinib was restarted.
[fig] Figure 1: PET -CT demonstrating an area of increased uptake in the left paravertebral area. [/fig]
[fig] Figure 2: MRI scan (T2 image) demonstrating the lesion that corresponds to the area of increased PET uptake. [/fig]
|
Hepatitis B Virus Infection in Human Immunodeficiency Virus Infected Southern African Adults: Occult or Overt – That Is the Question
Hepatitis B virus (HBV) and human immunodeficiency virus (HIV) share transmission routes and are endemic in sub-Saharan Africa. The objective of the present study was to use the Taormina definition of occult HBV infection, together with stringent amplification conditions, to determine the prevalence and characteristics of HBV infection in antiretroviral treatment (ART)naïve HIV +ve adults in a rural cohort in South Africa. The presence of HBV serological markers was determined by enzyme linked immunoassay (ELISA) tests. HBV DNA-positivity was determined by polymerase chain reaction (PCR) of at least two of three different regions of the HBV genome. HBV viral loads were determined by real-time PCR. Liver fibrosis was determined using the aspartate aminotransferase-to-platelet ratio index. Of the 298 participants, 231 (77.5%) showed at least one HBV marker, with 53.7% HBV DNA 2ve (resolved) and 23.8% HBV DNA +ve (current) [8.7% HBsAg +ve : 15.1% HBsAg 2ve ]. Only the total number of sexual partners distinguished HBV DNA +ve and HBV DNA 2ve participants, implicating sexual transmission of HBV and/or HIV. It is plausible that sexual transmission of HBV and/or HIV may result in a new HBV infection, superinfection and re-activation as a consequence of immunesuppression. Three HBsAg 2ve HBV DNA +ve participants had HBV viral loads ,200 IU/ml and were therefore true occult HBV infections. The majority of HBsAg 2ve HBV DNA +ve participants did not differ from HBsAg +ve HBV DNA +ve (overt) participants in terms of HBV viral loads, ALT levels or frequency of liver fibrosis. Close to a quarter of HIV +ve participants were HBV DNA +ve , of which the majority were HBsAg 2ve and were only detected using nucleic acid testing. Detection of HBsAg 2ve HBV DNA +ve subjects is advisable considering they were clinically indistinguishable from HBsAg +ve HBV DNA +ve individuals and should not be overlooked, especially if lamivudine is included in the ART.
# Introduction
Hepatitis B virus (HBV) and human immunodeficiency virus (HIV) share transmission routes and represent the two most important blood-borne pathogens in terms of prevalence, morbidity and mortality in sub-Saharan Africa, where both viruses are endemic. Of the 33.3 million adults and children living with HIV globally, 22.5 million reside in sub-Saharan Africa. Moreover, it is estimated that 65% to 98% of populations in sub-Saharan Africa have been exposed to HBV and 8% to 20% are chronic carriers of HBV [bib_ref] Epidemiology of hepatitis B virus in Africa, its genotypes and clinical associations..., Kramvis [/bib_ref] , far exceeding the 4% to 6% lifetime exposure rates and 0.2% to 0.5% carrier rates in regions of low endemicity. Thus, widespread co-infections are likely to occur, with 16% to 98% of HIV +ve individuals in sub-Saharan Africa being carriers of HBV or showing exposure to HBV [bib_ref] Hepatitis B virus and human immunodeficiency virus co-infection in sub-Saharan Africa: a..., Burnett [/bib_ref].
The progression of chronic HBV to cirrhosis, end-stage liver disease (ESLD), and hepatocellular carcinoma (HCC) is more rapid in HIV +ve individuals than those with HBV alone [bib_ref] Hepatitis C and B viruses: the new opportunists in HIV infection, Chung [/bib_ref] , with a significant increase in hepatic-related mortality rates [bib_ref] HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter..., Thio [/bib_ref]. Furthermore, HBV co-infection negatively impacts on HIV outcomes [bib_ref] Hepatitis B Virus Coinfection Negatively Impacts HIV Outcomes in HIV Seroconverters, Chun [/bib_ref].
Before the introduction of antiretroviral therapy (ART), the majority of HBV/HIV co-infected individuals were more likely to die from the clinical consequences of HIV than those of HBV [bib_ref] Hepatitis B virus and human immunodeficiency virus co-infection in sub-Saharan Africa: a..., Burnett [/bib_ref]. However, since the introduction of ART, the disease profile has changed, with increases in the proportion of mortality attributed to HBV-associated ESLD [bib_ref] Growing importance of liver disease in HIV-infected persons, Thomas [/bib_ref]. Thus, HBV/HIV co-infection can potentially impact on the safety and effectiveness of ART, requiring an integrated approach for the appropriate management of co-infected individuals [bib_ref] Confronting chronic hepatitis B virus infection in HIV: new diagnostic tools and..., Soriano [/bib_ref].
There is a paucity of comprehensive and standardized data describing HBV/HIV co-infection from southern African countries, where HIV prevalence is extremely high. Existing data show large discrepancies, with exposure rate to HBV in HIV +ve South Africans varying from 28% to 99.8% and HBsAg prevalence ranging from 0.4% to 23% [bib_ref] The prevalence of hepatitis B infection in a rural South African HIV..., Boyles [/bib_ref] [bib_ref] The prevalence of hepatitis B co-infection in a South African urban government..., Firnhaber [/bib_ref] [bib_ref] Occult hepatitis B virus infection in patients with isolated core antibody and..., Firnhaber [/bib_ref] [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] [bib_ref] Impact of lamivudine on HIV and hepatitis B virus-related outcomes in HIV/hepatitis..., Matthews [/bib_ref] [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref]. Differences can be attributed to different locations, study designs, laboratory measures and/or the composition of the study populations.
HIV infection has been implicated as a risk factor for the development of occult HBV infection (OBI) [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] , defined by the Taormina expert panel as the ''Presence of HBV DNA in liver (with detectable or undetectable HBV DNA in the serum) of individuals testing HBsAg negative by currently available assays. When detectable, the amount of HBV DNA in the serum is usually very low (,200 IU/ml)'' [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref]. Because liver biopsies are not commonly available, especially in resource-limited environments, OBI is usually detected by the analysis of sera [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref]. Furthermore, the experts differentiate between true occult (HBV viral load ,200 IU ml 21 ) and false occult where HBV DNA levels are comparable to those detected in HBsAg +ve infection (overt) and are usually as a result of infection by HBV variants with S gene escape mutants, producing HBsAg that is not recognized by detection assays [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref]. The clinical implications of OBI are unclear.
The prevalence of OBI in HIV infected individuals varies depending on the definition used, the sensitivity of the assay and the HBV viral loads [bib_ref] Occult hepatitis B virus infection in patients with isolated core antibody and..., Firnhaber [/bib_ref] [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref]. Furthermore, studies performed outside Africa, in areas of low HBV and HIV endemicity, cannot necessarily be extrapolated to Africa because of differences in host factors, epidemiology, transmission patterns and genotypes of the viruses between the two regions.
The objective of the present study was to use the Taormina definition of OBI [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] , together with stringent amplification conditions, to determine the prevalence and characteristics of HBV infection in ART-naïve HIV +ve adults entering a rural cohort in Mpumalanga Province, which has a HIV prevalence of 15.4%. No in-depth studies have been undertaken to determine the prevalence and characteristics of HBV/HIV coinfection in this province.
# Materials and methods
## Subjects
A new rural cohort was established at Shongwe Hospital in Mpumalanga Province in South Africa and 298 ART-naïve, HIV +ve adults were enrolled from July to November 2009. All had qualified for ART according to the then-current South African ART guidelines (CD4 counts ,200 cells mm 23 )and were recruited while undergoing treatment-readiness counselling. Universal HBV vaccination at 6, 10, and 14 weeks of age was introduced into the South African Expanded Programme on Immunization (EPI) in 1995 and therefore none of the participants were likely to have received this vaccination and self-reported as unvaccinated. Clinical and demographic data (including ALT levels, CD4 T-cell count, age, sex, height and weight) were obtained from hospital records, the National Health Laboratory Services (NHLS) databases and the TherapyEdge-HIV (TE) TM electronic patient record. All participants signed informed consent. The study was approved by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand and Mpumalanga Department of Health Research Ethics Committee.
## Serology
The presence of HBsAg, anti-HBsAg and anti-HBcAg was determined for 298 sera using the Monolisa TM HBsAg ULTRA, HBsAb ULTRA and HBcAb PLUS ELISA kits (Bio-Rad, Hercules, CA), respectively. HBeAg and anti-HBe tests were performed on HBV DNA +ve sera using the Monolisa TM HBeAg-Ab PLUS kit. Anti-HBcAg IgM was determined for 17 anti-HBc +ve HBV DNA +ve samples for which serum was available using the ARCHITECTH kit (Abbott Diagnostics, Wiesbaden, Germany). The M30-ApoptosenseH ELISA (Peviva AB, Stockholm, Sweden) was used on all sera to quantify the apoptosisassociated cytokeratin 18Asp396 neo-epitope as a measure of hepatocyte apoptosis [bib_ref] Immunocytochemical detection and mapping of a cytokeratin 18 neo-epitope exposed during early..., Leers [/bib_ref].
## Measurement of liver fibrosis
The aspartate aminotransferase (AST)-to-platelet ratio index (APRI) = (AST[/ULN]*100)/platelet count [10 9 L 21 ], a noninvasive measure of liver fibrosis in patients with chronic HBV [bib_ref] A simple noninvasive index (APRI) predicts advanced liver fibrosis in children with..., Lebensztejn [/bib_ref] , was calculated for 163 subjects for whom AST levels and platelet counts were available. APRI indicates liver fibrosis only when liver disease has reached a severely advanced stage, with significant fibrosis defined as APRI$1.5, and no fibrosis as APRI#0.5 [bib_ref] Prevalence, risk factors, and outcomes for occult hepatitis B virus infection among..., Lo Re [/bib_ref].
## Polymerase chain reaction (pcr)
DNA was extracted from 200 ml blood plasma with the QIAamp DNA Blood Mini Kit (QIAGEN Gmbh, Hilden, Germany) and eluted into 75 ml of best-quality water (BQW). Known positive and negative sera and BQW were used as controls for the extraction. Three regions of the HBV genome were amplified in a MyCycler TM thermocycler (Bio-Rad, Hercules, Ca, USA) using Promega Taq DNA polymerase (Promega, Madison, WI) . To avoid cross-contamination and false positives, the precautions and procedures of Kwok and Higuchi [bib_ref] Avoiding false positives with PCR, Kwok [/bib_ref] were strictly adhered to. DNA extraction, PCR, and electrophoresis were performed in physically separated venues.
## Real-time pcr quantification of hbv dna
PCR primers, HBV-Taq1 and HBV-Taq2 covering a region of the S gene (321 to 401 from the EcoRI site) with a FAM/TAMRA labelled TaqMan BS-1 probe [bib_ref] Sensitive and accurate quantitation of hepatitis B virus DNA using a kinetic..., Weinberger [/bib_ref] were used to quantify HBV DNA in an ABI 7500 Real Time PCR System (Applied Biosystems, Foster City, Ca, USA). A serial dilution of cloned plasmid DNA containing a single genome of HBV DNA, with concentrations ranging from 2610 1 to 2610 11 IU ml 21 , was used as template to generate the standard curve. The second WHO International Standard for HBV Nucleic Acid Amplification Techniques (product code 97/750 National Institute for Biological Standards and Controls (NIBSC); Hertfordshire, UK), which has a final concentration of 10 6 IU ml 21 was used as the internal standard. The standard curve, blank, positive and negative controls, and samples were all tested in duplicate. The measured IU/ml for each reaction was calculated using the Ct (cycle threshold) value of each PCR interpolated against the linear regression of the standard curve. The lower detection limit of our assay is ,20 IU ml 21 . The conversion formula of IU = copies/4.7 was used [bib_ref] Occult hepatitis B virus infection in patients with isolated core antibody and..., Firnhaber [/bib_ref] [bib_ref] Prevalence of hepatitis B virus (HBV) co-infection in HBV serologically-negative South African..., Firnhaber [/bib_ref].
# Statistical analysis
Clinical data were inspected visually. As all continuous variables showed a skewed distribution, the Mann-Whitney U test (Wilcoxon rank-sum test) was used to compare samples. Chisquared and Fisher's exact test were used to compare categorical variables. Exhaustive multivariate logistic regression analyses were performed. The R statistical language was used throughout.
# Results
## Serological and nucleic acid testing for hbv
The study group consisted of 298 adults (114 men and 184 women) with median age, CD4 count and BMI of 34 years, 147 cells mm 23 and 22 kg m 22 , respectively. Men were older than women and had lower CD4 counts .
The 298 participants were classified into five serogroups: 28 (9.4%) HBsAg +ve , 57 (19.1%) isolated anti-HBc +ve , 123(41.3%) anti-HBc +ve anti-HBs +ve , 11 (3.7%) anti-HBs +ve alone and 79 (26.5%) serologically 2ve for HBV. Six percent of men (7/114) were anti-HBs +ve alone compared to 2% (4/184) of women (p,0.05). The HBV serologically 2ve participants were significant-ly younger than most HBV serologically +ve groups and had significantly fewer lifetime sexual partners than those with isolated anti-HBs +ve (p,0.05). There was no significant difference in serologically-negative and -positive individuals in terms of CD4 counts, age of sexual debut, BMI, ALT and ApoptosenseH levels. Only five participants were HBeAg +ve and they did not differ from HBeAg 2ve individuals in either demographic or clinical features.
Screening for HBV DNA was carried out using primers targeting three non-overlapping regions of the HBV genome Of the entire group of 298, 26 (8.7%)/28 (9.4%) HBsAg +ve participants were HBV DNA +ve and together with the 45 (15.1%) HBsAg 2ve HBV DNA +ve participants were classified into 6 serogroups [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref]. Within the HBsAg 2ve groups, the frequency of HBV DNA was significantly higher in anti-HBc +ve alone individuals (16/57; 28.1%) compared to those anti-HBc +ve anti-HBs +ve (17/123; 13.8%) (p,0.05). The relative risk of an HBsAg 2ve individual, who was anti-HBc +ve alone, being HBV DNA +ve was twice as high as that of one with anti-HBc +ve anti-HBs +ve . The frequency of HBV DNA in the serologically 2ve group was not significantly different to that in the anti-HBc +ve alone or anti-HBc +ve anti-HBs +ve . Moreover, HBV DNA was not detected in any of the 11 isolated anti-HBs +ve individuals. Sufficient serum was available to test for anti-HBc IgM in 17 of 57 anti-HBc +ve HBV DNA +ve participants and all tested negative. Only three HBsAg 2ve HBV DNA +ve participants had viral loads ,200 IU ml 21 , thus meeting the Taormina criterion for true OBI [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref]. These participants had serological patterns of groups A, D and E, respectively [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref]. All other HBsAg 2ve HBV DNA +ve individuals had HBV viral loads .200 IU ml 21 .
Comparison of demographic and clinical characteristics between HBV DNA +ve and HBV DNA 2ve groups Visual inspection of plots and linear regression models of each of the continuous variables in Tables 2 and 3 (age, age at sexual debut, lifetime sexual partners, BMI (body mass index), ALT, Apoptosense, CD4 cell count, HBV viral load) against each other, for HBV DNA +ve versus HBV DNA 2ve , and HBsAg +ve versus HBsAg 2ve groups, revealed no significant correlation.
A multiple logistic regression model was used to determine predictors of HBV DNA positivity. In this model, only ALT levels were significant when all variables were included (p,0.05; OR = 1.01; 95% CI: 1.002-1.020). When the data were split according to gender, number of lifetime sexual partners was the only predictor in the females (p,0.05; OR = 1.16; 95% CI: 1.01-1.36) and ALT in the males (p,0.05; OR = 1.02; 95% CI: 1.004-1.030).
As shown in , the only variable that differentiated the HBV DNA +ve and HBV DNA 2ve groups was number of lifetime sexual partners (p,0.05). Regardless of whether they were HBV DNA +ve or HBV DNA 2ve , males were older than females, had a higher ALT and lower CD4 count. In the whole cohort and the HBV DNA 2ve group, females had a higher BMI (p,0.05) and fewer sexual partners than males (p,0.05). These differences were not seen in the HBV DNA +ve group. The age of sexual debut was . PCR primers and cycling parameters used for amplification of the three regions of the HBV genome. significantly different only when comparing males and females in the whole cohort.
Comparison of demographic and clinical characteristics between HBsAg +ve HBV DNA +ve and HBsAg 2ve HBV DNA +ve groups Data from the HBV DNA +ve participants were examined by logistic regression for predictors of HBV DNA-positivity in the absence of HBsAg. Only increasing age was weakly significant. The female subset showed that age was a significant predictor (p,0.05; OR = 1.28; 95% CI: 1.06-1.72). No predictors in the male subset were significant.
In the HBsAg 2ve HBV DNA +ve group, men were older and had significantly lower CD4 cell counts compared to females (p,0.05). Although the difference in ALT levels between the HBsAg +ve HBV DNA +ve and HBsAg 2ve HBV DNA +ve groups did not reach statistical significance , individuals who were HBsAg +ve anti-HBc +ve HBV DNA +ve [group C] had significantly higher ALT levels compared to individuals who were either serologically 2ve HBV DNA +ve [group A] (p,0.05) or anti-HBc +ve HBV DNA +ve [group E] (p,0.05) [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref]. There was no significant difference between the HBsAg +ve and HBsAg 2ve DNA +ve groups when ALT levels were coded into binary groups: .29 U/L for males and .19 U/L for females. HBV viral loads did not differ significantly between HBsAg +ve and HBsAg 2ve groups .
## Measurement of liver fibrosis using apri score
Ten percent of 163 individuals, for which data were available, had elevated APRI scores ($1.5), representing advanced fibrosis: 7.94% (10/126) HBV DNA 2ve [5.3% (2/38) seronegative and 9.1% (8/88) seropositive] and 16.2% (6/37) HBV DNA +ve [26.7% (4/15) HBsAg +ve HBV DNA +ve and 9.1% (2/22) HBsAg 2ve HBV DNA +ve ]. The frequency of liver fibrosis was significantly higher in HBsAg +ve HBV DNA +ve individuals compared to seronegative HBV DNA 2ve ones (p,0.05), but not to seropositive HBV DNA 2ve ones (p = 0.07). There was no significant difference between the HBsAg +ve and HBsAg 2ve HBV DNA +ve groups.
# Discussion
In this group of 298 southern African ART-naïve HIV +ve individuals, 231 participants had at least one HBV marker, giving an overall exposure to HBV of 77.5%, comparable to that in HBV monoinfected individuals [bib_ref] Epidemiology of hepatitis B virus in Africa, its genotypes and clinical associations..., Kramvis [/bib_ref]. In addition, almost one quarter of the group was HBV DNA +ve [fig_ref] Figure 1: Serological and DNA markers for HBV detected in 71 of 298 HIV... [/fig_ref] of whom almost two thirds were HBsAg 2ve . Direct comparison with other South African ART-naïve HIV +ve cohorts is difficult because of the different markers were used to measure exposure. In Limpopo Province, exposure to HBV, measured by anti-HBc and/or anti-HBs positivity, was 28.2% in a rural cohort [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] and 39.2% in antinatal HIV +ve women [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref]. This differs from the 63% HBV exposure rate (measured by at least one marker: HBsAg, anti-HBs or anti-HBc) found in a rural-urban HIV +ve cohort in Limpopo [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] and the much higher exposure rate of 99.8% in hospitaladmitted HIV +ve patients [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref]. In Gauteng Province, a 47% exposure was seen in an urban HIV +ve cohort where ,15% were HBV-positive as follows: 4.8% HBsAg +ve [bib_ref] The prevalence of hepatitis B co-infection in a South African urban government..., Firnhaber [/bib_ref] , 7.6% anti-HBc +ve HBV DNA +ve [bib_ref] Occult hepatitis B virus infection in patients with isolated core antibody and..., Firnhaber [/bib_ref] and 2.4% serologically 2ve HBV DNA +ve [bib_ref] Prevalence of hepatitis B virus (HBV) co-infection in HBV serologically-negative South African..., Firnhaber [/bib_ref].
The 9.4% HBsAg prevalence was comparable to that reported for some HIV +ve South African cohorts: 6.2% in anti-natal women in Limpopo Province [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] ; 7.1% in rural Eastern Cape (6.6% in ART-treated versus 8.8% in ART-naïve, p.0.05) [bib_ref] The prevalence of hepatitis B infection in a rural South African HIV..., Boyles [/bib_ref] ; and 6% in a . 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] 16 [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] country-wide study of treatment-naïve HIV +ve military personnel and their family members [bib_ref] Impact of lamivudine on HIV and hepatitis B virus-related outcomes in HIV/hepatitis..., Matthews [/bib_ref]. On the other hand, the HBsAg prevalence was higher than the 0.4% in another rural cohort in Limpopo Province [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] , double the 4.8% in a Gauteng urban cohort [bib_ref] The prevalence of hepatitis B co-infection in a South African urban government..., Firnhaber [/bib_ref] , but lower than the 11.3% in hospital-admitted Limpopo Province patients [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] , the 19.7% in miners [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] and the 22.9% from a rural-urban cohort in Limpopo Province [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref]. This difference in HBsAg prevalence correlates with the variations reported in HBV monoinfected individuals from different locales [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Progress towards the comprehensive control of hepatitis B in Africa: a view..., Kew [/bib_ref] [bib_ref] Preimmunization epidemiology of hepatitis B virus infection in South African children, Vardas [/bib_ref]. Regardless of whether they were HBV DNA +ve or HBV DNA 2ve , males were older, had higher ALT levels and lower CD4 counts than females . These differences are because males tend to come for treatment later than females [bib_ref] Sexassociated differences in pre-antiretroviral therapy plasma HIV-1 RNA in diverse areas of..., Grinsztejn [/bib_ref]. In the cohort as a whole and in the HBV DNA 2ve group, males had significantly more partners than females, with BMI significantly lower. In the HBV DNA +ve group, these factors did not differ between the genders. The only factor differentiating the HBV DNA +ve versus HBV DNA 2ve participants was the number of lifetime sexual partners , suggesting sexual transmission of HBV and/or HIV. It is plausible this mode of transmission may result in a new HBV infection, superinfection and re-activation as a consequence of immunesuppression. Twenty percent of the 71 HBV DNA+ve participants had possible markers of recent infection: 12 serologically 2ve HBV DNA +ve and 2 HBsAg +ve HBV DNA +ve [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref]. Of the 17 anti-HBc +ve HBV DNA +ve sera tested for anti-HBc IgM, none were positive.
The HBV serology of the cohort, the high frequency of HBeAgnegativity, the absence of anti-HBc IgM and the relatively low HBV viral loads [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref] reflect the natural history of HBV infection in sub-Saharan Africa, where most individuals are infected at childhood by horizontal transmission [bib_ref] Epidemiology of hepatitis B virus in Africa, its genotypes and clinical associations..., Kramvis [/bib_ref]. This means that most individuals have been exposed to HBV, and are protected by anti-HBV antibodies, by the time they become sexually active and acquire HIV. All isolated anti-HBs +ve participants were HBV DNA 2ve and the presence of anti-HBs with anti-HBc reduced the risk of being HBV DNA +ve . None of the participants had received HBV vaccination.
The HBsAg prevalence in this HIV +ve cohort was not different to HIV 2ve cohorts [bib_ref] Epidemiology of hepatitis B virus in Africa, its genotypes and clinical associations..., Kramvis [/bib_ref] [bib_ref] Hepatitis B virus and human immunodeficiency virus co-infection in sub-Saharan Africa: a..., Burnett [/bib_ref]. This differs from observations in areas of low HBV and HIV endemicity, where HBV and HIV are acquired simultaneously and therefore HBsAg prevalence in HIV +ve individuals is significantly higher than in HIV 2ve individuals [bib_ref] Hepatitis B virus and human immunodeficiency virus co-infection in sub-Saharan Africa: a..., Burnett [/bib_ref]. Only four participants in the present study were HBsAg +ve alone: two were HBV DNA +ve , whereas the other two were HBV DNA 2ve , even after repeated attempts to amplify HBV DNA, possibly indicating low viral loads undetectable by PCR. This might reflect the process of natural HBsAg clearance [bib_ref] What can be revealed by extending the sensitivity of HBsAg detection to..., Togashi [/bib_ref]. Although immune suppression by HIV may lead to the HBsAg +ve anti-HBc 2ve profile [bib_ref] Immune suppression as the etiology of failure to detect anti-HBc antibodies in..., Avettand-Fenoel [/bib_ref] , this is unlikely in these two cases, considering that ,59% of the participants were anti-HBc +ve , with a third of these having isolated anti-HBc. Moreover, HIV +ve patients with CD4,100 cells mm 3 are more likely to have isolated anti-HBc [bib_ref] Factors associated with isolated anti-hepatitis B core antibody in HIV-positive patients: impact..., Sun [/bib_ref].
HBV DNA without HBsAg was detected in 15.1% of the participants [fig_ref] Figure 1: Serological and DNA markers for HBV detected in 71 of 298 HIV... [/fig_ref]. This is within the 8% to 18% range for South African HIV +ve cohorts but again direct comparison is complicated by differences in study design [bib_ref] Occult hepatitis B virus infection in patients with isolated core antibody and..., Firnhaber [/bib_ref] [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref]. Twelve participants were serologically 2ve HBV DNA +ve , which can occur before the appearance of HBsAg, in the preseroconversion phase (indicating a recent infection), or at the tail end of the infection [bib_ref] Hepatitis B virus infection and transfusion medicine: science and the occult, Hollinger [/bib_ref]. Anti-HBsAg seroconversion, in the presence or absence of anti-HBc, decreased the relative risk of being HBV DNA +ve in the . 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] 16 [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] 18 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] [bib_ref] Immunocytochemical detection and mapping of a cytokeratin 18 neo-epitope exposed during early..., Leers [/bib_ref] 17 [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 16 [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 17 [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 16 [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] 20 [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] [bib_ref] Impact of lamivudine on HIV and hepatitis B virus-related outcomes in HIV/hepatitis..., Matthews [/bib_ref] [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] [bib_ref] Immunocytochemical detection and mapping of a cytokeratin 18 neo-epitope exposed during early..., Leers [/bib_ref] [bib_ref] A simple noninvasive index (APRI) predicts advanced liver fibrosis in children with..., Lebensztejn [/bib_ref] [bib_ref] Prevalence, risk factors, and outcomes for occult hepatitis B virus infection among..., Lo Re [/bib_ref] [bib_ref] Avoiding false positives with PCR, Kwok [/bib_ref] [bib_ref] Sensitive and accurate quantitation of hepatitis B virus DNA using a kinetic..., Weinberger [/bib_ref] [bib_ref] Prevalence of hepatitis B virus (HBV) co-infection in HBV serologically-negative South African..., Firnhaber [/bib_ref] [bib_ref] Progress towards the comprehensive control of hepatitis B in Africa: a view..., Kew [/bib_ref] [bib_ref] Preimmunization epidemiology of hepatitis B virus infection in South African children, Vardas [/bib_ref] [bib_ref] Sexassociated differences in pre-antiretroviral therapy plasma HIV-1 RNA in diverse areas of..., Grinsztejn [/bib_ref] [bib_ref] What can be revealed by extending the sensitivity of HBsAg detection to..., Togashi [/bib_ref] 27 19 [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] [bib_ref] Impact of lamivudine on HIV and hepatitis B virus-related outcomes in HIV/hepatitis..., Matthews [/bib_ref] [bib_ref] Hepatitis B virus infection and response to antiretroviral therapy (ART) in a..., Hoffmann [/bib_ref] [bib_ref] Increased exposure to hepatitis B virus infection in HIV-positive South African antenatal..., Burnett [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref] [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref] [bib_ref] Immunocytochemical detection and mapping of a cytokeratin 18 neo-epitope exposed during early..., Leers [/bib_ref] [bib_ref] A simple noninvasive index (APRI) predicts advanced liver fibrosis in children with..., Lebensztejn [/bib_ref] [bib_ref] Prevalence, risk factors, and outcomes for occult hepatitis B virus infection among..., Lo Re [/bib_ref] [bib_ref] Avoiding false positives with PCR, Kwok [/bib_ref] HBsAg 2ve group. This agrees with findings in HBV monoinfected [bib_ref] Prevalence of and Risk Factors for Hepatitis B Viremia After Spontaneous Hepatitis..., Chu [/bib_ref] and in HBV/HIV coinfected individuals [bib_ref] Occult hepatitis B in persons infected with HIV is associated with low..., Stuart [/bib_ref]. There was no difference in the demographics of the HBV DNA +ve subjects, with and without HBsAg . In the presence of HBsAg, there was no difference between males and females, whereas in the absence of HBsAg, males were older and had lower CD4 counts than females. Thus older males with lower CD4 counts are more likely to be HBsAg 2ve HBV DNA +ve . Lower CD4 counts have been associated with HBsAg 2ve viremia regardless of gender [bib_ref] Occult hepatitis B in persons infected with HIV is associated with low..., Stuart [/bib_ref] , however the median CD4 counts in that study were relatively higher (316 cells mm 23 versus 147cells mm [bib_ref] A simple noninvasive index (APRI) predicts advanced liver fibrosis in children with..., Lebensztejn [/bib_ref] in the present study) [bib_ref] Occult hepatitis B in persons infected with HIV is associated with low..., Stuart [/bib_ref].
In agreement with other studies [bib_ref] Incidence and risk factors for weight loss during dual HIV/hepatitis C virus..., Lo Re [/bib_ref] [bib_ref] Occult hepatitis B in HIV-infected patients, Shire [/bib_ref] , there were similar ALT levels in HBsAg +ve and HBsAg 2ve HBV DNA +ve participants and between HBV DNA +ve and HBV DNA 2ve participants. The absence of transaminitis is as a result of the immunosuppressed state of the HIV +ve subjects. Immunosuppression causes HBV reactivation and can lead to high viremia without clinical manifestation [bib_ref] Occult hepatitis B virus infection: detection and significance, Gerlich [/bib_ref]. The APRI score was used to compare the frequency of liver fibrosis in the HBV +ve versus HBV 2ve participants. The frequency of liver fibrosis was significantly higher in HBsAg +ve HBV DNA +ve individuals compared to seronegative HBV DNA 2ve ones, but not relative to seropositive HBV DNA 2ve ones. It is intriguing that there was no difference in the frequency of liver fibrosis between HBV DNA +ve individuals, with and without HBsAg.
The reactivation of an infection, which originated in childhood, can explain why no significant difference was seen in the HBV viral loads between the HBsAg +ve and HBsAg 2ve participants , nor between the different serological groups [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref]. Following HIV infection, HBV can reactivate in anti-HBs +ve only individuals, with and without the reappearance of HBsAg [bib_ref] Clinical reactivation of hepatitis B in anti-HBs-positive patients with, Vento [/bib_ref]. Group F, which had the lowest CD4 count of ,100 cells mm 23 , and by inference was the most immunosuppressed, was HBsAg +ve anti-HBc +ve anti-HBs +ve HBV DNA +ve with a viral load .10 2 IU ml 21 [fig_ref] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV... [/fig_ref]. Spontaneous reverse seroconversion, where anti-HBs disappears and HBsAg reappears can also occur in the presence of CD4 counts ,200 cells mm 23 [bib_ref] Hepatitis B and human immunodeficiency virus coinfection, Thio [/bib_ref]. Although HBV viral loads have been shown to be higher in HBV +ve HIV +ve individuals compared to HBV +ve ones [bib_ref] Interactions between HIV and hepatitis B virus in homosexual men: effects on..., Gilson [/bib_ref] , the HBV viral loads detected in the present study were comparable to those detected in HBV mono-infected individuals [bib_ref] A casecontrol study for differences among hepatitis B virus infections of genotypes..., Tanaka [/bib_ref]. This is probably because the majority of individuals were infected with subgenotype A1 [bib_ref] Genotyping and molecular characterization of hepatitis B virus (HBV) from antiretroviral treatment..., Makondo [/bib_ref] , which is characterized by relatively low viral loads in monoinfected individuals compared to other genotypes or subgenotypes [bib_ref] A casecontrol study for differences among hepatitis B virus infections of genotypes..., Tanaka [/bib_ref].
Only three HBsAg 2ve HBV DNA +ve patients had HBV loads ,200 IU ml 21 , meeting the Taormina criterion for OBI. Thus the majority of HBsAg 2ve HBV DNA +ve would be classified as false ''occult'' [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref]. It is possible that immunosuppression precludes true occult HBV infection. Because the majority of HBsAg 2ve HBV DNA +ve (''occult'') participants did not differ from HBsAg +ve HBV DNA +ve (overt) participants in terms of viral loads, CD4 counts, ALT levels and frequency of liver fibrosis, it may be more accurate to refer to these HBV infections as HBsAg-covert (HBsAg-cryptic overt) instead of false ''occult'' [bib_ref] Statements from the Taormina expert meeting on occult hepatitis B virus infection, Raimondo [/bib_ref].
HIV infection was demonstrated to be a risk factor for HBsAg 2ve HBV infection [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] , and pre-S mutations preventing HBsAg secretion [bib_ref] Properties of hepatitis B virus pre-S1 deletion mutants, Melegari [/bib_ref] , 'a' determinant mutations leading to detection escape and overlapping polymerase mutations affecting replication, may be responsible for this. This possibility was investigated and is presented in a follow-up paper, where 12 of 13 HBV S region sequences, from HBsAg 2ve participants, had pre-S and/or S mutations [bib_ref] Genotyping and molecular characterization of hepatitis B virus (HBV) from antiretroviral treatment..., Makondo [/bib_ref]. Another possible explanation for HBsAg-negativity may be that HIV co-infection prevents HBsAg secretion, as shown in co-infected hepatic cell lines [bib_ref] Increased intrahepatic apoptosis but reduced immune activation in HIV-HBV co-infected patients with..., Iser [/bib_ref].
Despite the possible limitations of this study, including its crosssectional nature, the absence of HIV viral loads, no HBV monoinfected patients and patients with higher CD4 counts for comparison, a number of important conclusions can be reached. The number of lifetime sexual partners was the only factor differentiating HBV DNA +ve and HBV DNA 2ve infections, suggesting sexual transmission of HBV and/or HIV. HBV +ve HIV +ve individuals were found to have significantly higher lifetime sexual partners than HBV-monoinfected individuals [bib_ref] HBV/HIV co-infection: the dynamics of HBV in South African patients with AIDS, Mayaphi [/bib_ref]. HBV infection in HIV +ve individuals was predominantly HBsAg 2ve , which did not differ significantly from HBsAg +ve infections in terms of viral loads, CD4 counts, ALT levels and frequency of liver fibrosis.
The detection of HBV DNA in the absence of HBsAg in this and other South African studies [bib_ref] Occult hepatitis B virus infection in patients with isolated core antibody and..., Firnhaber [/bib_ref] [bib_ref] High risk of occult hepatitis B virus infection in HIV-positive patients from..., Mphahlele [/bib_ref] [bib_ref] Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS..., Lukhwareni [/bib_ref] [bib_ref] Presence of occult HBV, but near absence of active HBV and HCV..., Barth [/bib_ref] has important implications for the clinical management of HIV in sub-Saharan Africa, where the burden of HBV/HIV co-infection is disproportionately high (24% in this study). Although the World Health Organization recommends that ART be initiated in HBV/HIV co-infected individuals irrespective of CD4 count, in South Africa we face a number of challenges. The most recent South African guidelines recommend initiation of treatment of patients with CD4 counts ,350 cells mm [bib_ref] A simple noninvasive index (APRI) predicts advanced liver fibrosis in children with..., Lebensztejn [/bib_ref] and HBsAg testing if ALT levels exceed 100 U L 21 . Considering that the highest median ALT levels (IQR) of 30 (19-59) U L 21 were found in the HBsAg +ve HBV DNA +ve group , which also had the highest frequency of advanced fibrosis, this cut-off value is inappropriate. Moreover, 65% of the 71 participants, who were HBV +ve HIV +ve , lacked HBsAg and HBV could only be detected by nucleic acid testing, which is unaffordable in resource-limited environments. Although the clinical significance of HBsAg 2ve infection is under debate [bib_ref] What can be revealed by extending the sensitivity of HBsAg detection to..., Togashi [/bib_ref] , it is imperative that HBV/HIV co-infection is detected before ART initiation, especially because lamivudine remains in two of the three drug regimens currently provided by the South African government and HBV can develop resistance to lamivudine. To determine the clinical relevance of HBsAg-covert HBV infection in our setting, prospective studies following ART initiation are in progress.
[fig] Figure 1: Serological and DNA markers for HBV detected in 71 of 298 HIV +ve participants. Overt refers to HBsAg +ve and ''occult'' to HBsAg 2ve . According to the Taormina definition, false occult infections are HBsAg 2ve with HBV viral load (VL) $200 IU ml 21 and true occult infections are HBsAg 2ve with HBV VL ,200 IU ml 21 [19]. doi:10.1371/journal.pone.0045750.g001 [/fig]
[fig] Figure 2: Box and whisker plot of HBV viral loads of the 71 HBV DNA +ve participants separated into the six serological groups (A to F), interpreted according toHollinger (2008) with modifications[35]. ''n'' indicates the number of participants in each group. ALT and CD4 cell counts for each group are indicated in the table below the plot as ''Median (Interquartile Range)''. Viral loads and CD4 cell counts did not differ significantly between the six serological groups. The five HBeAg +ve participants belonged to serological group C. doi:10.1371/journal.pone.0045750.g002 [/fig]
|
Associations between Intrinsic Heart Rate, P Wave and QT Interval Durations and Pulse Wave Analysis in Patients with Hypertension and High Normal Blood Pressure
The present study aimed to explore the relationship between electrocardiographic (ECG) and pulse wave analysis variables in patients with hypertension (HT) and high normal blood pressure (HNBP). A total of 56 consecutive, middle-aged hypertensive and HNBP patients underwent pulse wave analysis and standard 12-lead ECG. Pulse wave velocity (PWV), heart rate, intrinsic heart rate (IHR), P wave and QT interval durations were as follows: 7.26 ± 0.69 m/s, 69 ± 11 beats/minute, 91 ± 3 beats/minute, 105 ± 22 mm and 409 ± 64 mm, respectively. Significant correlations were obtained between PWV and IHR and P wave duration, respectively, between early vascular aging (EVA) and P wave and QT interval durations, respectively. Linear regression analysis revealed significant associations between ECG and pulse wave analysis variables but multiple regression analysis revealed only IHR as an independent predictor of PWV, even after adjusting for blood pressure variables and therapy. Receiver-operating characteristic (ROC) curve analysis revealed P wave duration (area under curve (AUC) = 0.731; 95% CI: 0.569-0.893) as a predictor of pathological PWV, and P wave and QT interval durations were found as sensitive and specific predictors of EVA. ECG provides information about PWV and EVA in patients with HT and HNBP. IHR and P wave durations are independent predictors of PWV, and P wave and QT interval may predict EVA.Central and Eastern Europe are regions with a high cardiovascular burden[1]. Romania was considered a high-cardiovascular-risk country according to the European Guidelines [2]. Thus, prophylactic measures deserve special attention, as well as information beyond standard cardiovascular risk factors in early prediction of future cardiovascular events[3].Arterial stiffness, the consequence of arteriosclerosis and subclinical atherosclerosis, is involved in cardiovascular disease development and, if assessed, can improve risk prediction, especially in hypertensive and intermediate risk patients[4][5][6][7][8].Heart rate and electrocardiography, especially repolarization variables, including QT interval duration, can also provide useful information in cardiovascular risk assessment[9][10][11][12][13]. The links between increased resting heart rate and mortality include sympathetic overactivity, increased metabolic rate and systemic inflammation, present in several systemic conditions, as well as cardiovascular risk factors[12]. An accelerated heart rate enables low, oscillatory endothelial shear stress and increased tensile stress, inducing a proinflammatory phenotype, with elevated levels of adhesion molecules, promoting atherogenesis and arterial stiffness[14,15]. Intrinsic heart rate is the heart rate under the simultaneous beta-blockade and muscarinic receptor blockade with propranolol and atropine, respectively[16,17].Chronological age has some limitations when assessing cardiovascular risk[18]. Vascular age can be assessed considering arterial stiffness, and early vascular aging can be defined if one's vascular age increases faster than one's chronological age, related to the upper 10, 20 or 25% of the pulse wave velocity (PWV) distribution in the population[19].Arterial stiffness is characterized by a decrease of elastin content, as well as by an increased production and accumulation of collagen within the arterial wall[20]. Arterial stiffness, assessed by pulse wave analysis, especially pulse wave velocity, is increased in hypertensive patients, regardless of blood pressure levels, and is related to endothelial dysfunction, vascular remodeling and progression of blood pressure[21,22]. On the other hand, increased arterial stiffness has a major effect on systolic blood pressure, wave reflections and cardiovascular risk[22].Considering the importance of prophylactic measures in hypertensive patients and the limits of the classical cardiovascular risk factors in predicting cardiovascular events, new, simple, inexpensive surrogate biomarkers are needed. Our objective is to explore the relationship between electrocardiographic (ECG) and pulse wave analysis variables in patients with hypertension and high normal blood pressure. We hypothesized that ECG variables can reveal increased arterial stiffness and early vascular aging in patients with elevated blood pressure values.Materials and MethodsStudy PopulationThe study included 56 consecutive hypertensive and high normal blood pressure (HNBP) patients, recruited from the Military Hospital Timisoara in the period of July 2016-March 2017.Patients with essential hypertension and high normal blood pressure, aged between 18 and 55 years, were included. Essential hypertension and high normal blood pressure were diagnosed according to the criteria of the European Society of Cardiology[23]. The most important exclusion criteria were secondary hypertension, atrial fibrillation, diabetes mellitus, history of coronary heart disease, myocardial infarction, stroke, transient ischemic attack or peripheral arterial disease, systemic inflammatory processes, active infections, trauma and therapy with statins[24].
# Introduction
The patients underwent pulse wave analysis and standard 12-lead electrocardiogram (ECG). Additional data related to diagnosis and therapy of the patients were available from medical records. The investigations were performed independently by different skilled investigators, who were not aware of the patient's diagnosis and data.
## Mobil-o-graph
Pulse wave analysis, including pulse wave velocity (PWV), augmentation index and pressure (AI, AP), vascular age and central and peripheral blood pressure were assessed using a Mobil-O-Graph (IEM GmbH, Stolberg, Germany), a noninvasive, cuff-based, one site, validated device. Pulse wave velocity is the speed at which the arterial pulse wave propagates along an arterial segment. Augmentation pressure (AP) is the increment in aortic pressure above its first systolic shoulder. Augmentation index is the ratio between augmentation pressure and pulse pressure, a measure of the contribution of wave reflection on the arterial pressure waveform. The methodology and the significance of the variables were previously published. PWV and vascular age were estimated from the reconstructed aortic pulse waveform via mathematical models. Early vascular aging was considered when one's vascular age was higher than the chronological age.
## Standard 12-lead ecg
Standard ECG was recorded at a paper speed of 25 mm/s. Resting heart rate (HR), P wave (P), PR (PR-the time from the onset of the P wave to the start of the QRS complex) and QT interval duration (QT) were automatically measured. QT was corrected considering heart rate, according to the Bazett formula, resulting in heart rate corrected QT interval (QTc). Intrinsic heart rate (IHR = 118 − (0.57 × age)) and the difference between IHR and resting heart rate (DHR) were also calculated. The amplitude of the R wave in leads V5 (RV5) and V6 (RV6) and of the S wave in leads V1 (SV1) and V2 (SV2) were manually measured.
# Statistical analysis
Categorical data are given as numbers and percentages, continuous data as means ± standard deviation. Correlations (Bravais-Pearson's, Kendall's and Spearman's), univariate and multivariate linear regression analysis and receiver-operating characteristic curve (ROC) analysis were used as statistical methods. Analyses were performed using IBM SPSS base edition (IBM, Armonk, NY, USA). Normality testing of dependent variables was performed using Q-Q plots, skewness and kurtosis. A p < 0.05 was considered statistically significant. PASS 2019 (NCSS Statistical Software, Kaysville, UT, USA) was used to conduct a power analysis before the study to calculate the sample size needed for the present study.
# Results
The study included middle-aged patients (48 ± 6 years) with hypertension and high normal blood pressure, most of them male (57%).includes demographic data, the values measured or calculated for several electrocardiographic variables and pulse wave analysis results. A total of 21 patients (37.5%) had high normal blood pressure, 16 (28.6%) grade 1, 13 (23.2%) grade 2 and 6 (10.7%) grade 3 hypertension. Hypertension grades considered the guidelines of the European Society of Cardiology for the management of arterial hypertension. Therapy included angiotensin converting enzyme inhibitors, sartans, diuretics, beta blockers and calcium channel blockers. Therapy angiotensin converting enzyme inhibitors (30%) sartans (17.5%) diuretics (32.5%) beta blockers (32.5%) calcium channel blockers (7.5%) SBP-systolic blood pressure; DBP-diastolic blood pressure; MAP-mean arterial pressure; PP-pulse pressure; PVW-pulse wave velocity; AP-augmentation pressure; AI-augmentation index; EVA-early vascular aging; HR-resting heart rate; IHR-intrinsic heart rate; DHR-the difference between IHR and resting heart rate; P-P wave duration; PR-PR interval duration; QRS-duration of the QRS complex; QT-duration of the QT interval; QTc-heart rate corrected QT interval using the Bazett formula; RV5-amplitude of the R wave in lead V5; RV6-amplitude of the R wave in lead V6; SV1-amplitude of the S wave in lead V1; SV2-amplitude of the S wave in lead V2.
Normality testing of dependent variables using Q-Q plots, skewness and kurtosis revealed that the data were approximately normally distributed.
Correlations were calculated for all ECG and pulse wave analysis variables. Significant correlations were obtained just between pulse wave velocity (PWV) and intrinsic heart rate (IHR) and the difference between IHR and resting heart rate (DHR). P wave duration significantly correlated with augmentation pressure, and augmentation index with heart rate corrected QT interval duration (QTc) and the amplitude of the S wave in lead V1 (SV1), respectively. No significant correlations were obtained between PWV and resting heart rate. AI-augmentation index, AP-augmentation pressure, IHR-intrinsic heart rate, DHR-the difference between IHR and resting heart rate; P-P wave duration; QT-duration of the QT interval; QTc-heart rate corrected QT interval using the Bazett formula; SV1-amplitude of the S wave in lead V1.
## Correlated variables correlation coefficient
Linear regression analysis revealed significant associations of IHR with PWV and early vascular aging (EVA). A prolonged QTc interval (exceeding 450 ms: QTc450) was significantly associated with EVA and augmentation index (AI). The association EVA-long QTc lost its significance after adjusting for systolic blood pressure (SBP). Significant associations were also found between AI and S wave in V1. Multiple linear regression analysis revealed only IHR as an independent predictor of PWV, even after adjusting for blood pressure variables and heart rate and blood pressure-lowering therapy. IHR lost its significant association with EVA after adjusting for blood pressure variables (SBP, mean arterial pressure (MAP) and pulse pressure (PP)) and therapy. ROC curve analysis revealed P wave duration as a predictor of pathological pulse wave velocity (pPWV). P wave and QT interval durationswere found as sensitive and specific predictors of EVA.
# Discussion
The present study reveals significant correlations and associations between electrocardiographic biomarkers and pulse wave velocity and early vascular aging, respectively, in middle-aged patients with hypertension and high normal blood pressure. Intrinsic heart rate and P wave duration were found as independent predictors of pulse wave velocity, while QT interval and P wave durations as independent predictors of early vascular aging, confirming our hypothesis that ECG variables can reveal increased arterial stiffness and early vascular aging in patients with elevated blood pressure values.
The influence of heart rate on PWV measurements remains controversial, with conflicting results observed in both acute and epidemiological studies. A significant blood pressure independent association between heart rate and PWV was mentioned, significant only for subjects with increased aortic stiffness. A higher resting heart rate was independently associated with arterial stiffness in healthy Korean adults. Heart rate plays differential roles in the development of arterial stiffness and subclinical atherosclerosis during young adulthood. Heart rate related changes in arterial stiffness were attributed to the viscoelasticity of the arterial wall.
# Discussion
The present study reveals significant correlations and associations between electrocardiographic biomarkers and pulse wave velocity and early vascular aging, respectively, in middle-aged patients with hypertension and high normal blood pressure. Intrinsic heart rate and P wave duration were found as independent predictors of pulse wave velocity, while QT interval and P wave durations as independent predictors of early vascular aging, confirming our hypothesis that ECG variables can reveal increased arterial stiffness and early vascular aging in patients with elevated blood pressure values.
The influence of heart rate on PWV measurements remains controversial, with conflicting results observed in both acute and epidemiological studies. A significant blood pressure independent association between heart rate and PWV was mentioned, significant only for subjects with increased aortic stiffness. A higher resting heart rate was independently associated with arterial stiffness in healthy Korean adults. Heart rate plays differential roles in the development of arterial stiffness and subclinical atherosclerosis during young adulthood. Heart rate related changes in arterial stiffness were attributed to the viscoelasticity of the arterial wall. According to other authors, the increase of arterial stiffness with heart rate is considered as an adjustment of the arterial tree to enable optimization of its performance and is more important in arteries supplying end capillaries with high permeability and low reflection coefficients. A lower heart rate prolongs ejection time, increases overlapping of the forward and backward pressure waves by widening the forward arterial waveform. The present study does not reveal any relationship between resting heart rate and pulse wave velocity and early vascular aging but reports significant correlations and associations between intrinsic heart rate and PWV and EVA, respectively. Multiple regression analysis demonstrates the independent predictive value of IHR for PWV, even after controlling for blood pressure and therapy-variables that influence arterial stiffness. According to our knowledge, this is the first study reporting a relationship between intrinsic heart rate and arterial stiffness. The possible link between IHR and arterial stiffness could be, besides age, also exercise training, considering that IHR is lower with training. Training-induced bradycardia is related to intrinsic electrophysiological changes in the sinus node, with downregulation of ion channels, especially funny channel HCN4. On the other hand, training may also improve arterial stiffness.
Several other studies reported links between electrocardiographic and pulse wave analysis biomarkers. QTc interval, prolonged in patients with Systemic Lupus Erythematosus, was related to subclinical atherosclerosis, measured by carotid-femoral pulse-wave velocity, after controlling for age and hypertension. A study including 54 apparently healthy participants revealed significant correlations and associations between Tpeak-Tend interval, a measurement of dispersion of the terminal part of the repolarization and predictor of life-threatening ventricular arrhythmias, and both brachial and aortic augmentation index, pulse wave velocity and early vascular aging. The Nagahama Study Group reported the longer corrected QT interval as an independent determinant of increased augmentation index and smaller pulse pressure amplification in persons free from cardiovascular symptoms and not receiving insulin therapy. The possible mechanisms explaining the association of impaired pulse wave and repolarization variables are electrophysiological remodeling related to aging, including action potential prolongation, blunted ionic exchanges across sarcolemma, increased myocardial fibrosis, increased ventricular load or subendocardial ischemia due to microvascular coronary atherosclerosis. A longer ventricular conduction period broadens the forward arterial pressure wave and increases the overlapping of the forward and backward pressure wave, which might explain the elevated augmentation index, which is also a marker of arterial stiffness, in patients with a long QT interval. The present study reveals the QT interval as a sensitive and specific predictor of EVA. P wave duration was previously correlated with arterial stiffness variables, related to interatrial block, in overweight patients. Body mass index was 27.14 ± 6.01 kg/m 2 , with a high prevalence of overweight (25%) and obesity (25%), which might be associated with a sedentary lifestyle in our study, but P wave duration exceeded 120 ms only in 7 patients (13%). The possible link could also be left atrial volume. A significant relationship between carotid arterial stiffness and left atrial volume was reported previously in patients with untreated hypertension. Ruaengsri et al. demonstrated in a canine model of mitral regurgitation that an increase of left atrial size is associated not only with an increased atrial fibrillation inducibility but also with a decrease in left ventricular ejection fraction, with an increase of end-systolic and end-diastolic volume. A study using the Campania Salute Network registry demonstrated that left atrial size was an independent predictor of cardiovascular events, regardless of other coexisting signs of target organ damage. Other factors that must be considered in the aging heart include atrial fibrosis associated with atrial conduction slowing and atrial fibrillation. Our study reveals P wave duration as an independent predictor of both EVA and pathological PWV.
Study limitations: arterial stiffness and early vascular aging are surrogate endpoints for cardiovascular events but using them demonstrates the prophylactic character of our study. The cross-sectional study design does not assure causality between the evaluated variables. The sample size is relatively small but the power analysis demonstrated an appropriate number of patients. Another limitation is related to the selection of study participants, including hypertensive patients receiving antihypertensive and heart rate-lowering drugs, which may question the validity of some results. However, the results did not change after controlling for blood pressure and heart rate-lowering therapy in the multiple regression analysis. The study lacks echocardiographic parameters and information about physical activity, which were not available for the patients included in the study but will be the aim of a future study. Mobil-O-Graph estimates pulse waves using an oscillometric method, which differs from the most widely applied applanation tonometry. Previous validation studies in different study populations, including hypertensive patients, revealed acceptable agreement between Mobil-O-Graph derived variables and invasive and noninvasive measurements and a good reproducibility of the technique.
Strengths of the study lie in the recruitment of real-world, middle-aged participants with grade 1-3 hypertension and high normal blood pressure, with a PWV increased for age in 21.4% of the participants and with both high and low cardiovascular risk, providing insights in terms of heart-vascular coupling rather than substitution of arterial stiffness measurement by ECG technique. As far as we know, our study is the first one reporting a significant relationship between intrinsic heart rate and arterial stiffness, respectively, as well as between P wave duration and EVA. The clinical implications of the present study are related to the prognostic importance of increased arterial stiffness and early arterial aging for cardiovascular risk. Simple, cost-effective, reproducible, quick to determine biomarkers, such as intrinsic heart rate, an index of sinus node function regardless of autonomic control, the duration of the P wave and QT interval can provide valuable, noninvasive information related to large vessels, arterial age and cardiovascular risk stratification in treated hypertensive patients, enabling personalized diagnosis and treatment of each patient, providing further targets for strategies to improve clinical outcomes and enable selection of patients requiring more aggressive therapy. Larger studies are needed in hypertensive patients to confirm the findings of the present study, to demonstrate that aging related functional and structural changes in the heart and artery are simultaneous and to reveal any connection between interstitial fibrosis related to aging associated remodeling of the myocardium, collagen content in the sinus node and arterial wall. Further prospective studies must confirm the synergistic prognostic effect of pulse wave velocity, early vascular aging, intrinsic heart rate, P wave and QT interval durations in hypertensive patients.
# Conclusions
ECG provides information about pulse wave velocity and early vascular aging in patients with hypertension and high normal blood pressure. IHR and P wave duration are independent predictors of pulse wave velocity in patients with hypertension and high normal blood pressure. P wave and QT interval duration are sensitive and specific predictors of EVA.
# Acknowledgments:
The data used to support the findings of this study are available from the corresponding author upon request.
## Conflicts of interest:
The authors declare no conflict of interest. A part of the data was presented at the Romanian Pathophysiology Conference 2019 and published as an abstract in Acta Medica Marisiensis. The founding sponsors did not influence study design, collection, analysis, or interpretation of data or drafting and submission of the manuscript. |
Prolonged Serum Alanine Aminotransferase Elevation Associated with Isotretinoin Administration
Isotretinoin is a highly effective oral retinoid derivative for severe forms of acne. Despite its high margin of safety, isotretinoin carries a risk of teratogenicity and mild to massive elevations of serum cholesterol and triglyceride levels, as well as infrequent transaminitis. Liver dysfunction induced by isotretinoin is rare but it poses a management dilemma. We describe a 16-year-old male in whom alanine aminotransferase (ALT) rose from a baseline of 13 to 288 U/L after 20 weeks of treatment with 1.0-1.4 mg/kg of oral isotretinoin daily. Though the patient remained asymptomatic, ALT levels did not return to normal limits for approximately 8 months after discontinuation of therapy, an observation that has not been documented in the literature. When oral isotretinoin was readministered for intractable facial acne 3 years later, liver enzymes remained normal throughout the course of therapy. Although the pathogenesis and prognosis of retinoid-induced hepatotoxicity are unknown, this case illustrates that isotretinoin may be safely readministered after normalization of liver function tests.
# Introduction
Isotretinoin is a highly effective oral retinoid therapy for severe and recalcitrant forms of nodular and cystic acne. Although isotretinoin has a high margin of safety, adverse effects include teratogenicity and elevated serum cholesterol, triglyceride, and transaminase levels. Rare hepatic dysfunction during long-term isotretinoin therapy has not caused liver damage on serial biopsies. Nonetheless, transaminase elevations have been reported as high as 5-fold above baseline normal values. We report a 16-yearold male who experienced a rise in serum alanine aminotransferase (ALT) from 13 to 288 U/L during isotretinoin therapy that did not return to baseline for approximately 8 months after cessation of the drug, a phenomenon not previously described in the literature. When intractable cystic acne developed in the same patient three years later, a second 5-month course of isotretinoin was accompanied by normal ALT levels and other parameters of liver function.
## Case presentation
An otherwise healthy 16-year-old boy presented with painful and disfiguring facial acne of several years duration unresponsive to a wide range of topical and oral antibiotic regimens. Physical examination showed extensive, red nodular, and fluctuant cysts with scarring on the face. Baseline blood tests were all within normal limits prior to starting isotretinoin, 30 mg twice a day (1.0 mg/kg), arrow 1). After two months of therapy, refractory acne prompted an increase of isotretinoin to 40 mg twice a day (1.4 mg/kg). The significance of an elevated ALT to 62 U/L [nl: <30 U/L] was uncertain, arrow 2). By the third month of isotretinoin therapy, there was a dramatic improvement of facial acne; however, ALT levels rose to 175 U/L. Isotretinoin Screening for hepatitis A, B, and C, as well as antinuclear antibodies, was negative. Considering enterohepatic recycling of potential hepatotoxic isotretinoin metabolites, the patient was started on cholestyramine powder, 4g daily with breakfast. This was discontinued after 7 weeks as ALT levels were returning towards normal, arrow 5). The ALT levels completely returned to normal values one month later and returned to baseline 2 months thereafter. Because of worsening cystic acne unresponsive to topical retinoids, topical antibiotics, and oral doxycycline, another course of isotretinoin was started approximately 3 years later. During the ensuing 5 months of treatment, ALT and other liver function tests remained normal. All inflammatory and cystic acne resolved by the conclusion of treatment.
# Discussion
Oral isotretinoin is a highly effective therapy for severe cystic acne. Since its approval by the US Food and Drug Administration in 1982, an estimated 12 million individuals have been treated with this drug. Its efficacy derives from serving as a ligand for the retinoic acid receptor (RAR), a nuclear receptor that forms a heterodimer with the retinoid X receptor (RXR), activating the transcription of target genes. The resulting decrease in sebocyte proliferation suppresses acne by reducing sebaceous gland size and sebum production.
Although isotretinoin therapy has a high margin of safety, side effects and complications are well characterized, with the most significant being teratogenicity. Other common side effects are dry skin, eyes, lips, and mucous membranes. Though altered LDL, HDL, and triglyceride levels are also common, the exact mechanism is unknown, and laboratory abnormalities spontaneously resolve after cessation of isotretinoin. Less frequent gastrointestinal side effects include nausea and the potential exacerbation of inflammatory bowel disease, though the latter remains controversial.
Hepatotoxicity in cases of hypervitaminosis A has been well documented. By contrast, hepatotoxicity associated with isotretinoin is uncommon. A recent study of highdose isotretinoin therapy in 80 patients with acne vulgaris revealed minimal, asymptomatic transaminase elevations in 23% of the cohort. While mild liver enzyme elevations during treatment with isotretinoin have been extensively reported, there is a comparable return to baseline (normal) after discontinuation of treatment. Our case is the first to describe persistent ALT elevations after discontinuation of isotretinoin therapy.
The mechanism of isotretinoin-induced hepatotoxicity is obscure, but postulated triggers include the parent compound itself or a toxic metabolite. Isotretinoin is a hydrophobic molecule that is highly bound (>99%) to plasma proteinsand metabolized by cytochrome P450 enzymes in the liver. Its serum half-life is approximately 14 hours, and the half-life of its major metabolite, 4-oxoisotretinoin, is closer to 28 hours. Additional metabolites were suggested by one study that followed the administration of an oral dose of 14 C-isotretinoin in which the serum halflife of labeled 14 C averaged 90 hours, indicating the presence of other isotretinoin metabolites. By contrast, when 14 C-isotretinoin was administered to two patients with ttube drainage of bile, there was a trend towards shortened half-lives, suggesting enterohepatic circulation of isotretinoin metabolites. Our case is unique as the sequestering agent, cholestyramine, was administered to reduce isotretinoin toxicity, associated with the potential accumulation of toxic metabolites recycling in the enterohepatic system. Serum ALT levels subsequently dropped by almost 50% within 2 weeks of starting cholestyramine. The association between normalized liver function tests (LFTs) and cholestyramine therapy remains uncertain; however, ALT levels no longer decreased after discontinuation of cholestyramine 2 months later.
Interpretation of lab abnormalities in the setting of isotretinoin therapy is challenging. In this case, elevated LFT levels may have been due to the high daily dose of isotretinoin. However, studies have shown that liver abnormalities associated with high-dose isotretinoin therapy are rare, occurring in 1.3% of patients. Modest serum ALT elevations associated with isotretinoin therapy are usually self-limited and generally do not require dose modification or discontinuation of therapy. Nonetheless, discontinuation of isotretinoin is recommended when ALT elevations are 3-fold or more above the upper limit of normal, as in our patient. Readministration of isotretinoin following asymptomatic hepatic function lab abnormalities poses a dilemma because there is no evidence-based standard of care. This case demonstrates that isotretinoin may be safely readministered in the context of intractable and disfiguring acne years after isotretinoin-induced lab abnormalities. Following normalization of biochemical parameters, etretinate may be another treatment option since cross-toxicity is not apparent between the various retinoid derivatives. Because of prolonged fat retention, etretinate therapy in women of childbearing potential is limited.
In summary, this case is the first to describe prolonged liver function test abnormalities following use and discontinuation of isotretinoin for acne. Isotretinoin should be withdrawn if ALT elevations are greater than 3 times the normal value. Cholestyramine may expedite normalization of lab values. Finally, this case suggests that readministration of isotretinoin, with close monitoring, after normalization of liver function tests may be a safe option for treatment resistant acne.
## Consent
The authors attest that the patient has given consent for the publication of this case.
# Disclosure
No identifying information, including names or images, was used in this publication. |
Advances in the modulation of ROS and transdermal administration for anti-psoriatic nanotherapies
Reactive oxygen species (ROS) at supraphysiological concentration have a determinate role in contributing to immuno-metabolic disorders in the epithelial immune microenvironment (EIME) of psoriatic lesions. With an exclusive focus on the gene-oxidative stress environment interaction in the EIME, a comprehensive strategy based on ROSregulating nanomedicines is greatly anticipated to become the mainstay of anti-psoriasis treatment. This potential therapeutic modality could inhibit the acceleration of psoriasis via remodeling the redox equilibrium and reshaping the EIME. Herein, we present a marked overview of the current progress in the pathomechanisms of psoriasis, with particular concerns on the potential pathogenic role of ROS, which significantly dysregulates redox metabolism of keratinocytes (KCs) and skin-resident or -infiltrating cells. Meanwhile, the emergence of versatile nanomaterial-guided evolution for transdermal drug delivery has been attractive for the percutaneous administration of antipsoriatic therapies in recent years. We emphasize the underlying molecular mechanism of ROS-based nanoreactors for improved therapeutic outcomes against psoriasis and summarize up-to-date progress relating to the advantages and limitations of nanotherapeutic application for transdermal administration, as well as update an insight into potential future directions for nanotherapies in ROS-related skin diseases.
# Introduction
Psoriasis (Ps) is a multifaceted disease related to chronic dysimmunity and genetic disease, which manifests in skin symptoms of demarcated erythematous and scaly lesions, accompanied by other systemic inflammatory comorbidities, like psychological illness, metabolic disturbance, arthritis, and cardiovascular disorders [bib_ref] Pathophysiology, clinical presentation, and treatment of psoriasis: a review, Armstrong [/bib_ref]. It has been affecting appropriately 125 million people worldwide [bib_ref] A systematic review of worldwide epidemiology of psoriasis, Michalek [/bib_ref] [bib_ref] Ashcroft DM. National, regional, and worldwide epidemiology of psoriasis: systematic analysis and..., Parisi [/bib_ref] , in which the age group of 60-69 years is recognized as a weighty psoriasis burden according to the Global Burden of Disease (GBD) 2019 study [bib_ref] The global, regional, and national burden of psoriasis: results and insights from..., Damiani [/bib_ref]. According to the clinical features, psoriasis is classified into cutaneous psoriasis and systemic psoriasis. Among the variants in cutaneous psoriasis, plaque psoriasis, also known as psoriasis vulgaris, is the most common phenotype, affecting ∼85-90% of patients with psoriasis [bib_ref] Cutaneous and systemic psoriasis: classifications and classification for the distinction, Yan [/bib_ref]. The histopathological feature of psoriatic lesions is parakeratosis in the thickened stratum corneum, the remarkably thickened epidermis with elongations into the dermis, and an abundance of different immune cells from dermis infiltration into the epidermis. Numerous studies have currently revealed that the direct or indirect cross-talking among different cell types in epithelial immune niches, plays a vital role in the pathogenesis of psoriasis and predominately emphasized the trigger role of oxidative stress in these cell types dysfunctions. Oxidative metabolites, namely reactive species, such as ROS/RNS, including superoxide anion hydroxyl radical (- OH − ), radical (- O 2 − ), hydrogen peroxide (H 2 O 2 ), singlet molecular oxygen ( 1 O 2 ), as well as nitric oxide, hydrogen sulfide, and oxidized lipids, prominently originates from mitochondrial electron transport chain (ETC), NADPH oxidases, other oxidases like peroxisome, several superoxide dismutases (SOD1-SOD3) and so on [bib_ref] Reactive oxygen species (ROS) as pleiotropic physiological signalling agents, Sies [/bib_ref] [bib_ref] Targeting free radicals in oxidative stress-related human diseases, Poprac [/bib_ref] [bib_ref] Acute activation of oxidative pentose phosphate pathway as first-line response to oxidative..., Kuehne [/bib_ref] [bib_ref] Interplay between DNA repair and inflammation, and the link to cancer, Kidane [/bib_ref]. The physiological concentration of reactive species is significant to orchestrate cellular redox signaling and guarantee diverse normal cell processes. Inversely, the supraphysiological level of these metabolites has the opposite pleiotropy. Therefore, it is imperative to deeply understand the role of detrimental ROS in the dyshomeostasis of keratinocytes (KCs) and immune cells in the epithelial immune microenvironment (EIME), ultimately leading to the generation and perpetuation of the inflamed cascade reaction in psoriasis [bib_ref] Reactive oxygen species (ROS) as pleiotropic physiological signalling agents, Sies [/bib_ref].
The conventional medications for psoriasis such as corticosteroids, vitamin D derivatives, targeting biologics, folic acid antagonists and calcineurin inhibitors are failing far to fulfill the current clinical need due to the systemic adverse reaction and the lower drug penetration [bib_ref] Multi-component clobetasol-loaded monolithic lipid-polymer hybrid nanoparticles ameliorate imiquimod-induced psoriasis-like skin inflammation in..., Pukale [/bib_ref] [bib_ref] Hyaluronic acid-based dissolving microneedle patch loaded with methotrexate for improved treatment of..., Du [/bib_ref]. Over the past decades, we have witnessed great success in medical nanomaterial, which has provided more and more nano-drugs and possible solutions for transdermal administration to improve psoriasis. The application of biomaterials to locally deliver conventional medications for psoriasis therapy can achieve an enhanced local drug concentration and circumvent system adverse reactions. Among the various nanotechnologies, several nanomaterials, e.g., microneedle and hydrogel, have demonstrated to be promising in clinical applications which are already on the market. In this review, we stay organized around the following two topics: firstly, we review how specific ROS perturbs and reprograms redox signaling pathways in KCs and immune cells, as well as provide a comprehensive understanding value of ROS as a promising therapeutic target for the applications in the treatment of psoriasis. In the end, we summarize the state-of-the-art ROS-regulating nano-medicines and nanomaterial-based therapies with distinctive transdermal delivery features used in anti-psoriatic therapies.
## Oxidative stress and its roles in different cell types dysfunctions of psoriasis
As the outermost immune and barrier organ of the human body, the skin is most vulnerable to be attacked by external insults, such as pathogen, toxication, pollution, trauma, UV rays, etc., concomitantly with an increased baleful ROS, consequently disturbing cutaneous defense mechanism and priming skin immune responses maintained by EIME [bib_ref] The epithelial immune microenvironment (EIME) in atopic dermatitis and psoriasis, Dainichi [/bib_ref] , which is composed of cellular communications among KCs, skin-resident and skin-infiltrating immune cells via interactions with a gradient of various chemo-attractants, such as chemokines, cytokines, vesicles and exosomes in the epidermis and papillary dermis [bib_ref] The epithelial immune microenvironment (EIME) in atopic dermatitis and psoriasis, Dainichi [/bib_ref] [bib_ref] Cytokinocytes: the diverse contribution of keratinocytes to immune responses in skin, Jiang [/bib_ref] , as shown as in . In the past decades, a dramatic increase in the numbers of evidence has highlighted that turbulence of EIME evokes the initiation and chronic inflammation in dermatoses significantly associated with oxidative stress [bib_ref] Oxidative stress and its role in skin disease, Trouba [/bib_ref] [bib_ref] NADPH oxidase inhibition rescues keratinocytes from elevated oxidative stress in a 2D..., Emmert [/bib_ref] [bib_ref] Oxidative stress as an important contributor to the pathogenesis of psoriasis, Pleńkowska [/bib_ref]. In addition to direct skin abnormality, systemic-based perturbations of metabolome also have appreciable effects on the pathogenesis of psoriatic inflammation [bib_ref] The metabolomics of psoriatic disease, Yan [/bib_ref]. As the pathogenic roles of increased oxidative stress, proinflammatory cytokines, adipocytokines, endoplasmic reticulum (ER) stress unbalance, and gut microbiota dysbiosis in the development of psoriasis with metabolic comorbidities are decoded, thus evaluating the metabolite profiles of psoriasis contributes to indicating biomarkers or novel therapeutic targets for Dysfunctional different cell types (KCs, skin-resident and -infiltrating immune cells function) mediate the propagation of inflammatory loops in EIME of psoriasis: turbulence of EIME evokes the initiation and chronic inflammation in psoriasis significantly associated with oxidative stress. Deleterious reactive metabolites ROS have a harmful role in inducing irreversible damage to these cells in EIME, thereby reprogramming their metabolic pathways involved in the development, proliferation, activation and function. Subsequently, intricately interwoven effects among these cells form clusters of inflammatory circuits in the pathophysiological EIME of cutaneous inflammation, ultimately giving rise to psoriasis prognosis and monitor response to the treatment [bib_ref] Cutaneous and systemic psoriasis: classifications and classification for the distinction, Yan [/bib_ref] [bib_ref] Metabolic syndrome and psoriasis: mechanisms and future directions, Hao [/bib_ref]. What's more, numerous discoveries exploring the pathogenic mechanism of psoriasis have shed light on intricately interwoven effects among keratinocyte, innate and adaptive immune cells to form clusters termed inducible skin-associated lymphoid tissue (iSALT) [bib_ref] The immunological anatomy of the skin, Kabashima [/bib_ref] [bib_ref] Skin-associated lymphoid tissues (SALT): Origins and functions, Streilein [/bib_ref] [bib_ref] Circuits and signals of the skin-associated lymphoid tissues (SALT), Streilein [/bib_ref] [bib_ref] Perivascular dendritic macrophages as immunobiological constituents of the human dermal microvascular unit, Sontheimer [/bib_ref] [bib_ref] Immunology of psoriasis, Lowes [/bib_ref] in the pathophysiological EIME of cutaneous inflammation, especially in psoriasis [bib_ref] Cytokinocytes: the diverse contribution of keratinocytes to immune responses in skin, Jiang [/bib_ref] [bib_ref] Amelioration of epidermal hyperplasia by TNF inhibition is associated with reduced Th17..., Zaba [/bib_ref]. Deleterious reactive metabolites like ROS have a harmful role in inducing DNA mutations, epigenetic alterations, post-translational modifications of protein kinase (cysteine residues) [bib_ref] Interplay between DNA repair and inflammation, and the link to cancer, Kidane [/bib_ref] , lipid peroxidation, and other key cellular components irreversible damage to these cells, thereby reprogramming their metabolic pathways of development, proliferation, activation and function, ultimately giving rise to psoriasis [bib_ref] Oxidative stress and its role in skin disease, Trouba [/bib_ref] [bib_ref] Oxidative stress as an important contributor to the pathogenesis of psoriasis, Plenkowska [/bib_ref] [bib_ref] Emerging roles of redoxmediated angiogenesis and oxidative stress in dermatoses, Xian [/bib_ref]. Therefore, disturbances in the oxidant-antioxidant system of the skin bring a dominant role in the pathogenesis of psoriasis [bib_ref] Reactive oxygen species in tumor necrosis factor-alpha-activated primary human keratinocytes: implications for..., Young [/bib_ref] , and keeping the dynamic equilibrium of the redox system is the most significant factor to sustain a myriad of normal biological processes in cells of EIME. Intracellular sophisticated antioxidative systems can counteract oxidative stressinduced ROS compounds, maintain redox homeostasis with a physiological threshold of ROS, and protect cells from an oxidative stress injury. The antioxidant capacity of the various skin cells is armed with the main cellular antioxidant pathways, including the main components of glutathione (GSH) pathways [bib_ref] Unearthing the secrets of mitochondrial ROS and glutathione in bioenergetics, Mailloux [/bib_ref] and transcriptional regulator NF-E2-related factor 2 (NRF2) [bib_ref] Unearthing the secrets of mitochondrial ROS and glutathione in bioenergetics, Mailloux [/bib_ref] [bib_ref] Redox regulation of immunometabolism, Muri [/bib_ref] [bib_ref] Natural Nrf2 modulators for skin protection, Boo [/bib_ref] , which translocate to the nucleus and binds to DNA promoters to initiate transcription of many antioxidant genes and cytoprotective proteins, to balance the level of oxidative metabolites, as shown as [fig_ref] Figure 2: ROS contributes to the rearranging immunometablism of different cell types, accompanied by... [/fig_ref]. Hence, we elucidate the focus role of ROS and molecular mechanism in skin KCs
## Oxidative stress-induced pathological signaling in kcs
It is admitted that KCs as amplifiers contribute to cellmediated psoriatic IL-23/IL-17 axis inflammation cascade effect in psoriasis. That is, cytokines, derived from IL23/IL-17 axis, induce ROS accumulation and cause redox dyshomeostasis of KCs, resulting in impairing the proliferation, differentiation and function of KCs via dysregulating phosphorylation/dephosphorylation key transcription factors and signal transductions in these cells, including NF-κB, STAT3, and others [bib_ref] Redox regulation of immunometabolism, Muri [/bib_ref] [bib_ref] Salidroside inhibits MAPK, NF-κB, and STAT3 pathways in psoriasis-associated oxidative stress via..., Xu [/bib_ref] [bib_ref] NF-kappaB inhibits TNF-induced accumulation of ROS that mediate prolonged MAPK activation and..., Sakon [/bib_ref]. These "activated" KCs exert a core pathogenic role in the cytokine-mediated various inflammation cascades [bib_ref] Human keratinocytes' response to injury upregulates CCL20 and other genes linking innate..., Kennedy-Crispin [/bib_ref] [bib_ref] Keratinocytes under fire of proinflammatory cytokines: bona fide innate immune cells involved..., Bernard [/bib_ref] [bib_ref] Tumor necrosis factor receptor signaling in keratinocytes triggers interleukin-24-dependent psoriasis-like skin inflammation..., Kumari [/bib_ref] , not merely serving as immune response triggers but also as proinflammatory non-immune cell effectors, which are capable of amplifying cytokine signal pathways from innate and adaptive immune cells to create a selfperpetuating autoimmune cytokine loop further so that persisting inward recruitment of leukocytes subsets into psoriatic lesions [bib_ref] Georgopoulos K. Direct control of regulatory T cells by keratinocytes, Kashiwagi [/bib_ref] [bib_ref] Skin immune sentinels in health and disease, Nestle [/bib_ref] , e.g., macrophages, neutrophils, myeloid DCs and T subsets. Young CN et al. found that the crucial psoriatic cytokine TNF-α could stimulate the activation of the mTOR-NF-κB pathway by ROS generation and ultimately production of inflammatory cytokines in KCs to initiate and maintain the progression of psoriasis; these ROS-induced cytokines could be attenuated by antioxidant enzyme and catalase, including taurine and N-acetyl-cysteine [bib_ref] Reactive oxygen species in tumor necrosis factor-alpha-activated primary human keratinocytes: implications for..., Young [/bib_ref]. Besides, rapamycin, an inhibitor of mTOR, could exert antiproliferative properties in the imiquimod (IMQ)-induced mice psoriasis via activating NRF2 signaling and restraining NOX2/4 from decreasing ROS generation [bib_ref] Rapamycin alleviates 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced aggravated dermatitis in mice with imiquimod-induced psoriasis-like dermatitis by..., Kim [/bib_ref]. Likewise, inhibiting the activity of NOX1/NOX4 in KCs could abrogate detrimental oxidative stress and rescue high levels of signature cytokines in a 2D model of atopic dermatitis and psoriasis [bib_ref] NADPH oxidase inhibition rescues keratinocytes from elevated oxidative stress in a 2D..., Emmert [/bib_ref]. CHF6001, a PD4 inhibitor, was reported to repress ROS through inactivating p47 (a subunit of the NOX complex 1) and then inhibit translocation of phosphorylated NF-κB, promoting the loss of cyclin D1 to alleviate redoxinflammatory crosstalk of psoriasis [bib_ref] Facchinetti F. The PDE4 inhibitor CHF6001 affects keratinocyte proliferation via cellular redox..., Woodby [/bib_ref]. Apart from NOX isoforms, dual oxidase 2 (DUOX2) homologues can also generate ROS. A study reported by Nadeem A et al. had shown that GPR43 agonists could activate epidermal GPR43-mediated DUOX2 and IL-6 signaling pathways to give rise to pernicious ROS, leading to Th17 polarization immune responses and deterioration of psoriasis [bib_ref] GPR43 activation enhances psoriasis-like inflammation through epidermal upregulation of IL-6 and dual..., Nadeem [/bib_ref]. Besides, Kumari S et al. uncovered that TNF-α induced the ROS-ERK pathway-dependent upregulation of IL-24 and activation of STAT3 signaling in stressed KCs upon KCs stimulated by endogenous and exogenous insults [bib_ref] Tumor necrosis factor receptor signaling in keratinocytes triggers interleukin-24-dependent psoriasis-like skin inflammation..., Kumari [/bib_ref]. STAT3, as an essential transcription factor, leads to the production of many cytokines in inflammatory processes of KCs [bib_ref] Adaptive and innate immunity in psoriasis and other inflammatory disorders, Schon [/bib_ref] [bib_ref] STAT3 links activated keratinocytes and immunocytes required for development of psoriasis in..., Sano [/bib_ref] , which in turn not only have an impact on disturbing the oxidant-antioxidant system but also recruiting a more deal of immune cells into the skin lesions to perpetuate a positive feedback inflammatory loop and remodeling extracellular matrix [bib_ref] Reactive oxygen species in tumor necrosis factor-alpha-activated primary human keratinocytes: implications for..., Young [/bib_ref] [bib_ref] Tumor necrosis factor receptor signaling in keratinocytes triggers interleukin-24-dependent psoriasis-like skin inflammation..., Kumari [/bib_ref]. Supraphysiological level of ROS makes the KCs be the state of 'oxidative distress' , which can induce the generation or modification of functional reductant protein networks under regulating the redox signaling pathways, as mentioned already, to control ROS production and availability [bib_ref] Reactive oxygen species (ROS) as pleiotropic physiological signalling agents, Sies [/bib_ref]. Among them, SIRT1, as a NAD-dependent deacetylase, plays a salient role in regulating the cellular pathological process of oxidative stress and autoimmune inflammation [bib_ref] Oxidative stress as an important contributor to the pathogenesis of psoriasis, Pleńkowska [/bib_ref] [bib_ref] Salidroside inhibits MAPK, NF-κB, and STAT3 pathways in psoriasis-associated oxidative stress via..., Xu [/bib_ref] [bib_ref] Sirt1: A potential therapeutic target in autoimmune diseases, Shen [/bib_ref] [bib_ref] Role of silent information regulator 1 (SIRT1) in regulating oxidative stress and..., Singh [/bib_ref]. In psoriasis, SIRT1 has been reported as a vital detoxifier of ROS-mediated redox signaling pathways, including MAPK, NF-κB, and STAT3, with downregulation of psoriatic inflammatory cytokines, suppression of keratinocyte hyperproliferation, and inhibition of angiogenesis [bib_ref] Salidroside inhibits MAPK, NF-κB, and STAT3 pathways in psoriasis-associated oxidative stress via..., Xu [/bib_ref] [bib_ref] Role of silent information regulator 1 (SIRT1) in regulating oxidative stress and..., Singh [/bib_ref] [bib_ref] Catalpol ameliorates psoriasis-like phenotypes via SIRT1 mediated suppression of NF-κB and MAPKs..., Liu [/bib_ref] [bib_ref] Chemerin/ChemR23 axis triggers an inflammatory response in keratinocytes through ROS-SIRT1-NF-κB signaling, Wang [/bib_ref] [bib_ref] Cimifugin ameliorates imiquimod-induced psoriasis by inhibiting oxidative stress and inflammation via NF-κB/MAPK..., Liu [/bib_ref] [bib_ref] Glycyrrhizin improves the pathogenesis of psoriasis partially through IL-17A and the SIRT1-STAT3..., Qiong [/bib_ref]. In addition, IL6/IL22-induced STAT3 activation in KCs was controlled by HO-1 induction and activation of protein tyrosine phosphatase SHP-1, accompanied by reduction of KCs hyperproliferation [bib_ref] Heme oxygenase-1 induction attenuates imiquimod-induced psoriasiform inflammation by negative regulation of STAT3..., Zhang [/bib_ref].
Similarly, the KEAP1/NRF2 system, as cytoprotective and antioxidative gene transcription, is critical in the redox signaling pathway with a core role in regulating inflammation, maintenance of epidermal differentiation and keratinization in response to ROS challenge [bib_ref] The KEAP1/NRF2 signaling pathway in keratinization, Ishitsuka [/bib_ref] [bib_ref] Therapeutic targeting of the NRF2 and KEAP1 partnership in chronic diseases, Cuadrado [/bib_ref]. The accumulated research has shown that a significant increase in detrimental ROS impairs the well-balanced cellular redox signaling pathways. It generates harmful protein oxidation products, leading to cell dysfunction and disease initiation. The expression of NRF2 is reduced and its downstream regulatory genes in psoriatic skin tissues. In the IMQ-induced psoriasis-like mice model, NRF2/HO-1 in the skin lesion was decreased. The accumulation of excessive ROS activated the NF-κB pathway, concomitantly with the secretion of proinflammatory cytokines IL-17, IL-23, IL-1β and VEGF expression [bib_ref] Galangin ameliorates Imiquimod-Induced psoriasis-like skin inflammation in BALB/c mice via down regulating..., Sangaraju [/bib_ref] [bib_ref] Astilbin reduces ROS accumulation and VEGF expression through Nrf2 in psoriasislike skin..., Wang [/bib_ref]. The reduction of other prototypical examples of redox signaling-mediated antioxidative enzymes is also involved in the pathomechanism of psoriasis, such as GSH, Px, CAT, and SOD [bib_ref] Salivary antioxidants and oxidative stress in psoriatic patients: can salivary total oxidant..., Skutnik-Radziszewska [/bib_ref] [bib_ref] Deciphering psoriasis. A bioinformatic approach, Melero [/bib_ref]. In addition, several aquaporins (AQP3, AQP8 and AQP9), referred to as 'peroxiporins' , facilitate the transportation of H 2 O 2 across cellular membranes to regulate downstream intracellular signalings [bib_ref] Aquaporin-3-mediated hydrogen peroxide transport is required for NF-κB signalling in keratinocytes and..., Hara-Chikuma [/bib_ref] [bib_ref] Aquaporin-3 mediates hydrogen peroxide uptake to regulate downstream intracellular signaling, Miller [/bib_ref]. The study of Hara-Chikuma M et al. demonstrated that AQP3-facilitated H 2 O 2 transport was the precondition of NF-κB activation in KCs participating in the acceleration of psoriasis; In AQP3 knockout mice AQP3 (-/-), IL-23-mediated psoriasiform skin inflammation was reduced [bib_ref] Aquaporin-3-mediated hydrogen peroxide transport is required for NF-κB signalling in keratinocytes and..., Hara-Chikuma [/bib_ref]. Taken together, the abovementioned studies of dysfunctional KCs suggest that oxidative stress-related signaling pathways make a difference in the pathogenesis of psoriasis, and it is worthy of decreasing cytokines gene expression and obstructing the autoimmune loop for the treatment of psoriasis effectively via quenching generation and traffic of triggers-induced pernicious ROS with ROS-depletion or -blockade approaches.
## Oxidative stress-mediated abnormal immunometabolism in immune cells of psoriasis the role of oxidative stress in macrophage dysfunction
It is well established that macrophages derived from monocytes lineage cells are the main component cells of innate immunity. Most human and animal studies have emphasized the crucial role of macrophages in the pathogenesis of psoriasis [bib_ref] Macrophages in dermatology: pathogenic roles and targeted therapeutics, Kuraitis [/bib_ref] [bib_ref] Activated macrophages are essential in a murine model for T cell-mediated chronic..., Wang [/bib_ref] [bib_ref] Recent insights into the immunopathogenesis of psoriasis provide new therapeutic opportunities, Nickoloff [/bib_ref]. ROS/RNS contributes to rearranging macrophage differentiation and exerting their effector functions in response to tissue environments via intermediating the main cellular oxidationreduction (redox) pathways, including glutathione (GSH) pathways, and NF-E2-related factor 2 (NRF2) [bib_ref] Redox regulation of immunometabolism, Muri [/bib_ref] [bib_ref] ROS play a critical role in the differentiation of alternatively activated macrophages..., Zhang [/bib_ref] [bib_ref] Mitochondria in the regulation of innate and adaptive immunity, Weinberg [/bib_ref]. Myeloid-derived suppressor cells (MDSCs) have been demonstrated involved in the progress of psoriasis. GSH synthesis in MDSCs isolated from the bone marrow of IMQ-induced psoriatic mice model with ROS accumulation was reduced, resulting in interruption of MDSCs differentiation into CD11c + MHC II + dendritic cells and CD206 + M2 macrophages to exacerbate skin inflammation [bib_ref] Acitretin promotes the differentiation of myeloid-derived suppressor cells in the treatment of..., Liu [/bib_ref]. In murine macrophages, LPS/ IMQ could induce ROS/RNS-NF-κB/ERK/JNK signaling pathway and decrease the expression of NRF2, increasing iNOS and other inflammatory cytokines to exacerbate psoriasiform skin inflammation [bib_ref] Protective effects of ambroxol in psoriasis like skin inflammation: exploration of possible..., Sunkari [/bib_ref]. It is admitted that the major endogenous enzymatic sources of O 2 and H 2 O 2 are transmembrane NADPH oxidases and NADPH oxidase 2 complexes (NOX2) complex-generated ROS can participate in regulating the metabolism and oxidation-reduction signaling pathways of macrophages and neutrophils involved in chronic inflammation, such as mannan-induced Ps and PsA (MIP), rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) [bib_ref] Reactive oxygen species (ROS) as pleiotropic physiological signalling agents, Sies [/bib_ref]. Zhong J et al. demonstrated that Nos2-derived NO modulated the pathogenic IL-1α secretion from the local macrophages, which acted to downstream target innate lymphoid cell 3 (ILC3), resulting in the up-regulation of IL-17 A to trigger and accelerate the development of MIP [bib_ref] Mannan-induced Nos2 in macrophages enhances IL-17-driven psoriatic arthritis by innate lymphocytes, Zhong [/bib_ref]. Moreover, mitochondria are also the source of cellular ROS [bib_ref] a mitochondrial love-hate triangle, Brookes [/bib_ref]. Once the antioxidant defense mechanism is compromised, the aggravation of mitochondrial malfunction-induced ROS could provoke the onset of chronic inflammatory diseases [bib_ref] Succinate dehydrogenase supports metabolic repurposing of mitochondria to drive inflammatory macrophages, Mills [/bib_ref] [bib_ref] Mitochondrial mutagenesis correlates with the local inflammatory environment in arthritis, Harty [/bib_ref]. Mitochondrial ROS is capable of NLRP3 inflammasome activation [bib_ref] Mitochondria in the regulation of innate and adaptive immunity, Weinberg [/bib_ref] [bib_ref] The inflammasomes, Schroder [/bib_ref] , which is a crucial reactor to trigger innate immune defenses by maturing proinflammatory cytokines such as interleukin (IL-1β and IL-18) [bib_ref] The inflammasomes, Schroder [/bib_ref] [bib_ref] A role for mitochondria in NLRP3 inflammasome activation, Zhou [/bib_ref]. In the peripheral blood of untreated patients with psoriasis, the expression levels of inflammasome sensors, IL-1β and IL-18 were enhanced; Verma D et al. demonstrated that TNF-α upregulated pro-IL-1β and pro-IL18 and stimulated these inflammasome activities via increasing ROS and activation of NLRP3 signaling pathways [bib_ref] Enhanced inflammasome activity in patients with psoriasis promotes systemic inflammation, Verma [/bib_ref]. A previous study reported that administration of propranolol (the nonselective beta-blocker) was relevant with exacerbation of psoriasis, ascribed to inhibition of autophagic flux, with an abundance of ROS-producing mitochondria in cutaneous LCs, leading to IL23A production [bib_ref] Lysosomotropic beta blockers induce oxidative stress and IL23A production in langerhans cells, Müller [/bib_ref]. Additionally, HO-1, considered an antioxidative enzyme, is responsible for cytoprotective molecules against oxidative damage and inflammation. Recent shreds of evidence have mentioned that drugs with the property of increased HO-1 expression are protective in animal models of psoriasis, such as curcumin, carnosol, DMF and hemin [bib_ref] Galangin ameliorates Imiquimod-Induced psoriasis-like skin inflammation in BALB/c mice via down regulating..., Sangaraju [/bib_ref] [bib_ref] Naturally derived heme-oxygenase 1 inducers attenuate inflammatory responses in human dendritic cells..., Campbell [/bib_ref] [bib_ref] Fumarates improve psoriasis and multiple sclerosis by inducing type II dendritic cells, Ghoreschi [/bib_ref]. Elevated HO-1 expression could attenuate psoriasiform inflammation via inhibiting iNOS in macrophages and maintaining DCs immune tolerance phenotypes [bib_ref] Mitochondrial mutagenesis correlates with the local inflammatory environment in arthritis, Harty [/bib_ref] [bib_ref] Naturally derived heme-oxygenase 1 inducers attenuate inflammatory responses in human dendritic cells..., Campbell [/bib_ref] [bib_ref] Heme oxygenase 1 attenuates the development of atopic dermatitis-like lesions in mice:..., Kirino [/bib_ref]. Oppositely, some conflict data suggested that variation of HO-1 system expression in macrophages not only presented beneficial roles, but detrimental outcomes in other diseases, such as cancer and infection [bib_ref] Heme oxygenase 1 controls early innate immune response of macrophages to Salmonella..., Mitterstiller [/bib_ref] [bib_ref] Heme oxygenase-1: emerging target of cancer therapy, Chau [/bib_ref]. Based on the abovementioned research, it should be realized that macrophages, as the main effector components of innate immunity, are activated by intrinsically and extrinsically oxidative stress through tissue-specific signals to promote the secretion of disease context-specific cytokines [bib_ref] Electrophilic properties of itaconate and derivatives regulate the IκBζ-ATF3 inflammatory axis, Bambouskova [/bib_ref] [bib_ref] Cis-khellactone inhibited the proinflammatory macrophages via promoting autophagy to ameliorate imiquimod-induced psoriasis, Feng [/bib_ref]. Therefore, the treatment of unspecific antioxidants could alleviate disease depending on the situation of specific pathogenesis. Furthermore, a full elucidation of oxidative stress in the pathogenesis and progression mechanisms of disease-specific is a precondition for their use as therapeutic antioxidants in medical applications. In psoriasis, proinflammatory macrophages are essential contributors to the pathophysiological inflammatory cascade by forming immunological clusters termed inducible skin-associated lymphoid tissue (iSALT) in the dermis of cutaneous inflammation [bib_ref] Cytokinocytes: the diverse contribution of keratinocytes to immune responses in skin, Jiang [/bib_ref] [bib_ref] Perivascular dendritic macrophages as immunobiological constituents of the human dermal microvascular unit, Sontheimer [/bib_ref] [bib_ref] Immunology of psoriasis, Lowes [/bib_ref] [bib_ref] Amelioration of epidermal hyperplasia by TNF inhibition is associated with reduced Th17..., Zaba [/bib_ref] [bib_ref] Perivascular leukocyte clusters are essential for efficient activation of effector T cells..., Natsuaki [/bib_ref] , which is indispensable for elicitation of adaptive immunity and ultimately orchestrated immune-related signal pathways in KCs, causing a switch into keratinocyte hyperplasia and aberrant differentiation in chronic psoriasiform skin inflammation [bib_ref] Activated macrophages are essential in a murine model for T cell-mediated chronic..., Wang [/bib_ref]. Thus, inhibition of the proinflammatory phenotypes of macrophages could be of therapeutic benefit in the psoriasis context. Emerging selective targets against oxidative stress of macrophages and skin inflammation in dermatologic diseases are given by the above multiple specific ROS-mediated signaling pathways and offer a perspective for better-refined redox medicine.
## The role of oxidative stress in neutrophil dysfunction
Psoriasis has a wide range of clinical subtypes, which are determined by complicated fine-tuning of innate and adaptive immune responses [bib_ref] Adaptive and innate immunity in psoriasis and other inflammatory disorders, Schon [/bib_ref]. General pustular psoriasis (GPP) is an acute and severe systemic inflammation characterized by neutrophilic-rich dysfunction, leading to sterile pustules in skin lesions. It was triggered by neutrophil extracellular traps (NETs) formation (termed as NETosis, a cell death process), which is implicated in autoimmune inflammatory reactions and induced by neutrophil activation and respiratory burst, to release the non-specific effects of CitH3, enzymatic proteins (like neutrophil elastase and MPO), cytosolic proteins (such as S100 calcium-binding proteins) and recruit pro-inflammatory immune cells [bib_ref] Cellular mechanisms of NETosis, Thiam [/bib_ref] [bib_ref] Neutrophil extracellular traps enriched in oxidized mitochondrial DNA are interferogenic and contribute..., Lood [/bib_ref] [bib_ref] Histone acetylation promotes neutrophil extracellular trap formation, Hamam [/bib_ref] [bib_ref] Cannabidiol modifies the formation of NETs in neutrophils of psoriatic patients, Wójcik [/bib_ref]. The process of NETosis mediated by reactive oxygen species (ROS)derived from mitochondria and NADPH oxidase could induce autoantibody production, resulting in uncontrolled inflammatory response and tissue pathology [bib_ref] A review of neutrophil extracellular traps (NETs) in disease: potential anti-NETs therapeutics, Mutua [/bib_ref]. In the onset of psoriasis, KCs are attacked and stressed upon various stimuli, such as trauma, drugs, and infections, followed by the release of damaged DNA/RNA, LL-37, AMPs, DAMPs and other cytokines/chemokines from these activated KCs, which could initiate innate immune responses and attract more neutrophils infiltration into the epidermis to form Munro or Kogoj abscesses, this sterile pustules constitutes typical pathological manifestations of GPP. Meanwhile, these stressed neutrophils produce weblike NETs under ROS-induced respiratory burst, and the release of MPO, elastase and hydrolase from NETs are known to transform inactive precursors of the IL-1β and IL-36 family released from KCs into more biologically active mature bodies, leading to the characteristic pro-inflammatory imbalance of the IL-36 autocrine and autoinflammatory circuits in generalized pustular psoriasis [bib_ref] Cannabidiol modifies the formation of NETs in neutrophils of psoriatic patients, Wójcik [/bib_ref] [bib_ref] Autoinflammatory psoriasis"-genetics and biology of pustular psoriasis, Uppala [/bib_ref] [bib_ref] Myeloperoxidase modulates inflammation in generalized pustular psoriasis and additional rare pustular skin..., Haskamp [/bib_ref]. In the meantime, activated neutrophils secrete psoriatic cytokines such as IL-17 A and IFN-γ members, which could aggravate the self-perpetuating autoimmune cytokine loop in KCs so that persisting inward recruitment of leukocytes subsets into psoriatic lesions and promotion of KCs proliferation [bib_ref] NF-kappaB inhibits TNF-induced accumulation of ROS that mediate prolonged MAPK activation and..., Sakon [/bib_ref] [bib_ref] Human keratinocytes' response to injury upregulates CCL20 and other genes linking innate..., Kennedy-Crispin [/bib_ref] [bib_ref] Keratinocytes under fire of proinflammatory cytokines: bona fide innate immune cells involved..., Bernard [/bib_ref]. There is mounting evidence of NETs formation at obvious risk of autoimmune diseases, an inflammatory neutrophil subset with characteristics of aged CD10 neg CD16 low CD11b low neutrophils appeared in lesional skin and circulation of psoriasis and these aged neutrophils increased IL-17 expression by T cells in a NETosis-dependent way [bib_ref] Immunomodulatory aged neutrophils are augmented in blood and skin of psoriasis patients, Rodriguez-Rosales [/bib_ref] ; immature CD10 neg CD16 neg CD11b neg/low neutrophils from patients detected a higher ROS level under TNF-α plus f-MLF stimulation as compared with those of healthy controls [bib_ref] Immunomodulatory aged neutrophils are augmented in blood and skin of psoriasis patients, Rodriguez-Rosales [/bib_ref]. Noting that the enzyme MPO is induced by exposure of neutrophils to various forms of oxidative stress, which is one of the important markers of NETosis [bib_ref] Cannabidiol modifies the formation of NETs in neutrophils of psoriatic patients, Wójcik [/bib_ref]. this pro-oxidative and pro-inflammatory hemeprotein is recognized to provide a preponderant role in NETs formation; MPO-deficient neutrophils from MPO-deficient individuals cumulatively associated with GPP, the formation of NETs was predominately reduced compared to healthy donors [bib_ref] Myeloperoxidase modulates inflammation in generalized pustular psoriasis and additional rare pustular skin..., Haskamp [/bib_ref]. Similarly, serum MPO levels displayed a significant increase and caused the injury of antioxidative defenses in psoriasis children [bib_ref] Plasma oxidation status and antioxidant capacity in psoriatic children, Bacchetti [/bib_ref]. Notably, in the IMQ-induced psoriatic mouse model, levels of MPO and oxidative stress were also upregulated [bib_ref] Assessment of an imiquimod-induced psoriatic mouse model in relation to oxidative stress, Baek [/bib_ref]. In combination, these accumulations of evidence supported that redox imbalance between oxidant-antioxidants occurred very early in neutrophils, thereby oxidative burst, activation and degranulation of neutrophils involved in the process of NETosis, which implicated in the prolonged persistence of neutrophils in the affected psoriatic individuals and the inability of resolvable inflammation. Conclusively, these data implicate that detrimental ROS contributes to the induction of NETs and the application of ROS-elimination drugs could restore the potential occurrence of NETs formation, thereby shifting the balance to predominant anti-inflammatory signals to counteracting many neutrophil-mediated diseases, in particular GPP. Therefore, targeted NETs degradation biological treatment may be conducive to the containment of sustained neutrophil-mediated skin inflammation.
## The role of oxidative stress in dc dysfunction
Much substantial evidence from clinical studies and experimental models has emphasized the critical role of DCs in the pathogenesis of autoimmune diseases, especially psoriasis [bib_ref] The role of dendritic cells in autoimmunity, Ganguly [/bib_ref]. The aberrant hyperactivation of DCs could bridge the innate and adaptive immune responses, sufficient to induce psoriasis. it is well appreciated that the cellular immunometabolism changes and redox signaling pathways of immune cells are tightly interwoven and interdependent to regulate their differentiation, proliferation and function [bib_ref] Redox regulation of immunometabolism, Muri [/bib_ref]. Mizuguchi S et al. unveiled that in a psoriatic mouse model, the suppression of mtROS attenuated the exacerbation of IMQ stimulation psoriasiform dermatitis and IMQ-induced DC activation in vitro was suppressed by inhibition of the generation of mtROS [bib_ref] Mitochondrial reactive oxygen species are essential for the development of psoriatic inflammation, Mizuguchi [/bib_ref]. A similar result, reported by Al-Harbi NO et al. that activation of BTK signaling in CD11c + DCs upregulated oxidative stress, associated with significant elevation of inflammatory mediators, which are crucial factors in the pathogenesis of IMQ-induced psoriasislike inflammation in mice [bib_ref] Therapeutic treatment with Ibrutinib attenuates imiquimod-induced psoriasis-like inflammation in mice through downregulation..., No [/bib_ref]. Asides from these data, the cellular redox disequilibrium of DCs could adversely affect their ability to induce activation of T-cells and regulate the polarity of the immune response via glutathione depletion interfering in DC maturation and IL-12 production [bib_ref] Glutathione depletion inhibits dendritic cell maturation and delayed-type hypersensitivity: implications for systemic..., Kim [/bib_ref]. As a consequence, these advances suggest that ROS homeostasis is inseparable from maintaining the well-balanced cellular immunometabolism of DCs. Potential therapeutic strategies by neutralizing the excess of ROS could open up new insight into prevention in psoriasis.
## The role of oxidative stress in t cell dysfunction
The pivotal role of T cells in the pathogenesis of psoriasis is evidenced by substantial studies. Dysfunctional different T cells subpopulations and their associated cytokines are crucially involved in the onset or exacerbation of psoriasis, and blockade of these cytokinemediated inflammations could be identified as potential therapeutic targets. Strikingly, dynamic cellular redox reactions are obbligato for ensuring and regulating the homeostatic maintenance of different T cells subsets differentiation and cellular functions. The disruption of redox homeostasis in T cell subsets provides susceptibility to numerous immunopathies [bib_ref] Redox regulation of immunometabolism, Muri [/bib_ref] [bib_ref] Intracellular free radical production by peripheral blood T lymphocytes from patients with..., Amico [/bib_ref]. Esmaeili B et al. demonstrated antioxidant defense mechanisms were disordered by elevated ROS in stimulated memory CD4 + T cells from psoriasis patients. It is well-known that regulatory T cells (Tregs) are regarded as protect effect on preventing psoriasis, and excessive ROS would reduce the ratio of Treg: Th17 cells by promoting the proliferation and differentiation of pro-inflammatory Th17/Th1/ Th22 cells and reversely suppression of the frequency of Treg to sustain the process of psoriasis [bib_ref] Proanthocyanidins: novel treatment for psoriasis that reduces oxidative stress and modulates Th17..., Lai [/bib_ref] [bib_ref] Anti-inflammatory activity of compounds isolated from Astragalus sinicus L. in cytokine-induced keratinocytes..., Kim [/bib_ref]. Furthermore, detrimental cellular ROS-induced oxidized 8-oxo-dGTP and DNA also could amplify Th17 subset cells, along with striking elevation of IL-17-producing γδ T cells in lymph nodes [bib_ref] MTH1 inhibitors for the treatment of psoriasis, Bivik Eding [/bib_ref]. Considering the essential role of the dermal IL-17-producing γδ T cells in psoriasis, its redox regulation engaged in immunometabolism gains more attention as the pivotal player in developing psoriasis [bib_ref] Pivotal role of dermal IL-17-producing γδ T cells in skin inflammation, Cai [/bib_ref]. Recent advances demonstrated that mTORC2 constrained mitoROS production in γδ T cells, causing impairment of γδ T17 differentiation, which is critical innate dermal predominate IL-17-producing cells in the development and aggravation of psoriasis [bib_ref] Roles of mTORC1 and mTORC2 in controlling γδ T1 and γδ T17..., Yang [/bib_ref]. These previous researches make us conscious that more efforts should be paid to comprehensively decipher the definite role of ROS mediated in metabolic rewiring and impaired functions of T cells in disease-specific pathogenesis. It conduces accelerating the discovery of more advanced treatment modalities to restore the balance of ROS levels in T cells for combating autoimmune diseases, particularly psoriasis.
## The role of oxidative stress in other immune cells dysfunction
Similar to what is discovered in the abovementioned immune cells involved in the occurrence of psoriasis, extensive research has been performed to detail the crucial role of skin-resident ILCs-associated cytokines IL-17 and IL-22 in contributing to driving dermal inflammation, particularly in psoriasis [bib_ref] Skin-resident innate lymphoid cells converge on a pathogenic effector state, Bielecki [/bib_ref] [bib_ref] Innate lymphoid cells: major players in inflammatory diseases, Ebbo [/bib_ref]. ILCs belong to a family of innate immune cells lacking antigen-specific receptors and are classified into three subgroups (ILC1, ILC2, and ILC3) according to their key transcription factors expression and cytokines production [bib_ref] Innate lymphoid cells: major players in inflammatory diseases, Ebbo [/bib_ref] [bib_ref] Innate lymphoid cells-how did we miss them?, Walker [/bib_ref]. Among them, type 3 ILCs (ILC3s) play a central role in the etiology and disease severity of psoriasis, which was ascribed to the elevated number of IL-22-and IL-17 A/Fproducing ILC3s induced by their expression of RORγt transcription factors in psoriatic skin and blood [bib_ref] Innate lymphoid cells: major players in inflammatory diseases, Ebbo [/bib_ref] [bib_ref] Composition of innate lymphoid cell subsets in the human skin: enrichment of..., Teunissen [/bib_ref] [bib_ref] Characterization of innate lymphoid cells in human skin and blood demonstrates increase..., Villanova [/bib_ref] [bib_ref] A new player on the psoriasis block: IL-17A-and IL-22-producing innate lymphoid cells, Ward [/bib_ref]. RORγt + ILC and γδ T cells are also prerequisites for driving psoriasiform plaque formation in the IMQ-induced disease models through the aggregation delivery of IL-17 A, IL-17 F, and IL-22 into the skin inflammation [bib_ref] Rorγt + innate lymphocytes and γδ T cells initiate psoriasiform plaque formation..., Pantelyushin [/bib_ref]. Similar to the immunometablism of other immune cells, ILC plasticity could be supervised by redox metabolic pathways and cytokine milieu. The deficiency of NOX2 shifted Tbet + ILC1s transdifferentiation into RORγt + ILC3s in a redox-dependent manner through IL-1β production and aggravated the inflamed joints of Ncf1 −/− mice [bib_ref] Increased ILC3s associated with higher levels of IL-1β aggravates inflammatory arthritis in..., Chan [/bib_ref]. Likewise, one study also found that Nos2-derived NO upregulated IL-17-producing ILC3 by IL-1α stimulation from the local macrophages participated in triggering and progressing the development of MIP. In addition to the better-studied pathogenesis of ILCs in psoriasis, contributions of NK cell-mediated innate immune responses to inflammatory skin diseases, especially psoriasis, have shown increasingly emerging [bib_ref] Natural killer cells in atopic and autoimmune diseases of the skin, Von Bubnoff [/bib_ref] [bib_ref] Innate lymphocytes in psoriasis, Polese [/bib_ref] [bib_ref] Role of innate immune cells in psoriasis, Sato [/bib_ref]. Different subsets of NK cells take part in dysregulating the imbalance of immune response to many autoimmune diseases through the induction of their cytokines and cytotoxic functions [bib_ref] The role of natural killer cells in autoimmune diseases, Kucuksezer [/bib_ref]. A study reported by Gilhar A et al. illuminated that NK and NKT cells from autologous human lymphocytes were injected into nonlesional skin grafts from psoriatic patients on mice could give rise to representative psoriatic skin inflammation with the expression of inflammatory epidermis signatures [bib_ref] Psoriasis is mediated by a cutaneous defect triggered by activated immunocytes: induction..., Gilhar [/bib_ref]. Besides, NKT cells with IFN-γ/CCR5 expression in psoriatic skin were relevant to the severity of psoriasiform hyperplasia and microabscess [bib_ref] Interferon-γ/ CCR5 expression in invariant natural killer T cells and CCL5 expression..., Kono [/bib_ref]. Certainly, analogous to the effect of redoxassociated metabolic pathways on ILC development and function, the probabilities of NK cell-fate transitions at different stages are also shifted upon autophagy perturbations-inducing ROS disequilibrium. The excessive ROS production under the condition of disrupting dysfunctional mitochondria elimination caused by the deletion of Atg5 or Atg7, severely compromised homeostasis and the maturity of NK cells. Additionally, progressive research in mast cells (MCs) enables satisfactory characterization of cells and their delicate roles in the complex network of psoriasis. Gaudenzio N et al. reported that IFN-γ-primed human MCs caused abundant immunologic synapses with CD4 + T cells, concomitantly with an enhancement of the production of Th22 and IL-22/IFN-γ-producing Th cells from the circulating memory CD4 + T-cell pool; a productive infiltration of IL-22 + CD4 + T cells observed in contact with mast cells in human psoriatic skin biopsies [bib_ref] Human mast cells drive memory CD4 + T cells toward an inflammatory..., Gaudenzio [/bib_ref]. Strikingly, the proportion of IL-22-producing mast cells occupied 20-80% in patients with psoriasis, and skin mast cells expressed IL-22 and IL-17 mRNA [bib_ref] Human mast cells are major IL-22 producers in patients with psoriasis and..., Mashiko [/bib_ref]. Furthermore, IL-24 from activated T cell-derived microvesicles motivated MCs and excessive MCs activation in psoriasis could produce IL-24, subsequently provoking STAT3 phosphorylation of KCs [bib_ref] T cellderived microvesicles induce mast cell production of IL-24: relevance to inflammatory..., Shefler [/bib_ref] [bib_ref] Immune cell infiltration analysis demonstrates excessive mast cell activation in psoriasis, Zhang [/bib_ref].
Advances in understanding MCs activation and degranulation have shown that the role of mitochondrial translocation and ROS involved in activating MCs of allergic inflammatory diseases is overwhelming [bib_ref] Mitochondrial uncoupling protein 2 inhibits mast cell activation and reduces histamine content, Tagen [/bib_ref] [bib_ref] Human mast cell degranulation and preformed TNF secretion require mitochondrial translocation to..., Zhang [/bib_ref] [bib_ref] Mitochondriatargeted antioxidant skq1 (10-(6´-plastoquinonyl)decyltriphenylphosphonium bromide) inhibits mast cell degranulation in vivo and..., Chelombitko [/bib_ref] [bib_ref] The role of reactive oxygen species and nitric oxide in mast cell-dependent..., Swindle [/bib_ref]. Skin biopsies from AD revealed that mitochondrial translocation was present in the degranulation and TNF secretion of human skin mast cells [bib_ref] Human mast cell degranulation and preformed TNF secretion require mitochondrial translocation to..., Zhang [/bib_ref]. However, the causal relationship between ROS-stimulated MCs activation and psoriasis is needed to be done to expand our basic knowledge. Overall, a disordered oxidant-antioxidant system, in combination with the turbulence of cellular ROS homeostasis from enhanced activation of redox signaling pathways, renders the disturbed immunometablism of immune cells particularly vulnerable to trigger and exacerbation of psoriasis. Comprehensively studying the pathophysiological role played by ROS in the abovementioned immune cells related to the pathogenesis of psoriasis would help to design potential dysfunctional effector cells-targeted anti-inflammatory and anti-psoriatic drugs.
## Therapeutic drugs targeting oxidative stress in eime of psoriasis
To date, the therapeutic efficacies of various agents depend on how well these cycles of inflammation mediated by the abovementioned dysfunctional cells in EIMEs of psoriasis are broken. In consideration of the aforementioned multi-faceted influences of oxidative stress present in the dysfunctional different cell types in EIME of psoriatic inflammation (summarized in , considerable research has demonstrated disorganized cellular redox signaling pathways in these dysfunctional cells whose induced multiple inflammatory networks could be sophisticatedly modulated and blocked by a variety of chemical agents or drugs. As shown in , DMF has been previously reported as a broad-spectrum antiinflammatory drug. It could be used to treat psoriasis via modulating the phenotypic switch of immune cell types through glutathione depletion and reprogramming the cellular redox balance, particularly the modulation of macrophages and type II dendritic cells [bib_ref] Fumarates improve psoriasis and multiple sclerosis by inducing type II dendritic cells, Ghoreschi [/bib_ref] [bib_ref] Dimethyl fumarate alters intracellular ca handling in immune cells by redox-mediated pleiotropic..., Herrmann [/bib_ref]. Alongside these mechanisms, DMF could also impair NETs formation in polymorphonuclear granulocytes isolated from psoriasis patients via limiting oxidative burst capacity, mediated by depletion of intracellular GSH levels [bib_ref] Dimethyl fumarate modulates neutrophil extracellular trap formation in a glutathione-and superoxide-dependent manner, Hoffmann [/bib_ref]. Building on a study reporting that DMF could cause short-term oxidative stress and activate the antioxidant signaling response of transcription factor NRF2, increasing the antioxidant protein expression and modulating cellular redox state to alter the expression of key genes or proteins related to calcium signaling of immune cell activation [bib_ref] Dimethyl fumarate alters intracellular ca handling in immune cells by redox-mediated pleiotropic..., Herrmann [/bib_ref]. In type II DCs, DMF performed its therapeutic effect via inducing glutathione (GSH) depletion of DCs, followed by increasing the expression of antioxidant hemoxygenase-1 (HO-1) gene and impaired phosphorylation of STAT1 to ameliorate psoriasis and MS (Multiple Sclerosis) [bib_ref] Fumarates improve psoriasis and multiple sclerosis by inducing type II dendritic cells, Ghoreschi [/bib_ref]. CBD (Cannabidiol), as a wide spectrum of antioxidant and anti-inflammatory modulators, is studied for application in preventing and treating redox imbalance and inflammation-associated diseases [bib_ref] A systematic review of cannabidiol dosing in clinical populations, Millar [/bib_ref] [bib_ref] Antioxidative and antiinflammatory properties of cannabidiol, Atalay [/bib_ref]. Indeed, CBD could be considered a potential anti-NETotic factor to inhibit NETosis formation by reducing NADPH oxidase and MPO expression [bib_ref] Cannabidiol modifies the formation of NETs in neutrophils of psoriatic patients, Wójcik [/bib_ref]. Ibrutinib, a BTK inhibitor, could ameliorate psoriasiform inflammation by attenuating ROS and inflammatory mediators in CD11c + DCs [bib_ref] Therapeutic treatment with Ibrutinib attenuates imiquimod-induced psoriasis-like inflammation in mice through downregulation..., No [/bib_ref]. Apremilast, a PDE4 inhibitor, improvement of intracellular cAMP, could augment IL-10-producing Bregs and its concomitant decrease in Th1 cells, IFNγ-producing NKT cells and IL-17-producing NKT cells and suppress IFNγ + CD3 + T cells and IL-17 + CD3 + T cells for combating PsA and Ps [bib_ref] The effect of apremilast on signal transduction and IL-10 production in CD39high..., Sakkas [/bib_ref] [bib_ref] Apremilast is a selective PDE4 inhibitor with regulatory effects on innate immunity, Schafer [/bib_ref] [bib_ref] A review in psoriasis and psoriatic arthritis, Keating [/bib_ref]. Other natural immunomodulatory compounds, such as curcumin [bib_ref] Naturally derived heme-oxygenase 1 inducers attenuate inflammatory responses in human dendritic cells..., Campbell [/bib_ref] , proanthocyanidins [bib_ref] Proanthocyanidins: novel treatment for psoriasis that reduces oxidative stress and modulates Th17..., Lai [/bib_ref] [bib_ref] Red-kerneled rice proanthocyanidin inhibits arachidonate 5-lipoxygenase and decreases psoriasis-like skin inflammation, Toda [/bib_ref] , and galanin [bib_ref] Galangin ameliorates Imiquimod-Induced psoriasis-like skin inflammation in BALB/c mice via down regulating..., Sangaraju [/bib_ref] perform their anti-proliferative and anti-inflammatory effects in different cell types via utilization of their important pharmacological properties of antioxidant to neutralize baleful ROS, interrupt pro-inflammatory MAPK, NF-κB, and STAT3 signalings and potentiate anti-inflammatory NRF-2, SIRT1, and HO-1 pathways. Other non-canonical anti-inflammatory drugs, like Ambroxol [bib_ref] Protective effects of ambroxol in psoriasis like skin inflammation: exploration of possible..., Sunkari [/bib_ref] and MTH1 inhibitors [bib_ref] MTH1 inhibitors for the treatment of psoriasis, Bivik Eding [/bib_ref] could be used as antipsoriatic drugs possessing capabilities of aiming at ROS elimination in specific diseasingcausing cell types to ameliorate psoriasis. In addition to the above-mentioned chemical and non-classical drugs as a potential treatment for psoriasis, some of the main The pathogenetic role of ROS in dysfunctional different cell types (KCs, skin-resident and -infiltrating immune cells functions) mediated propagation of inflammatory loops in the EIME of psoriasis classical traditional anti-psoriasis drugs also can regulate immune cell metabolism and keratinocyte excessive proliferation. For example, MTX, the classical anti-psoriasis drug [bib_ref] Treatments for psoriasis and the risk of malignancy, Patel [/bib_ref] , can also be regarded as an antioxidant, which can neutralize free radicals and reactive oxygen superoxide (O 2 − ), thereby inhibiting the formation of malondialdehyde acetaldehyde (MAA) adducts. Vitamin A is an indirect antioxidant that indirectly regulates many genes involved in mediating typical antioxidant responses and can prevent lipid peroxidation, thus remodeling metabolic pathways and gene expression profiles in tissues and cells [bib_ref] Vitamin A and vitamin E: will the real antioxidant please stand up?, Blaner [/bib_ref]. However, their traditional therapeutic routes targeting the abovementioned inflammatory network are still not satisfactory due to their substantial toxicity concerning internal organs, nonspecific targeting, low effective drug concentration of skin lesions, specific risks of infection, and poor patient compliance [bib_ref] Therapeutics targeting the IL-23 and IL-17 pathway in psoriasis, Ghoreschi [/bib_ref] [bib_ref] Calderon M. Nanocarriers for skin applications: where do we stand?, Tiwari [/bib_ref]. 90% of voters in the International eDelphi Consensus Meeting recommended switching the MTX route to subcutaneous administration against psoriasis for averting oral adverse events [bib_ref] International edelphi study to reach consensus on the methotrexate dosing regimen in..., Van Huizen [/bib_ref].
Topical therapy is the safe, convenient, and most widely used approach for the transdermal delivery of classical antipsoriatic drugs to treat mild psoriasis and consolidation treatment of moderate-to-severe psoriasis in current clinical applications. the circumvent of adverse reactions and sufficient concentration of therapeutic drug at the target lesion could be facilitated by transdermal administrations [bib_ref] Calderon M. Nanocarriers for skin applications: where do we stand?, Tiwari [/bib_ref] [bib_ref] Treatment of psoriasis: novel approaches to topical delivery, Wollina [/bib_ref]. A number of strategies for the transdermal delivery of bioactive drugs have been investigated for the clinic. Compared with the parenteral delivery route, topical different formulations [bib_ref] Novel topical nanocarriers for treatment of psoriasis: an overview, Dadwal [/bib_ref] , including ointment, cream, lotion, liquid, emulsions, gel formulations and nanomedicines-assisted transdermal delivery of drugs could directly repress the deterioration of psoriasis to achieve comparable therapeutic effects through a variety of The therapeutic effects of common natural compounds and drugs in the targeted regulation of ROS-mediated pathogenesis of psoriasis
## Mth1 inhibitors normalization of the neutrophils and t cells frequencies in the skin and skin-draining lymph nodes, decrease of il-17-producing γδ t cells and preventation of il-17-downstream genes in kcs
Ex-vivo psoriasis PBMC/ HEKn/Th17-driven inflammation in mice [bib_ref] MTH1 inhibitors for the treatment of psoriasis, Bivik Eding [/bib_ref] Astragalus sinicus L. Inhibition of NF-κB signaling cascades in cytokinestimulated KCs, and suppression of CD4 + T cells differentiated into Th2 and Th17 cell subsets via scavenging intracellular ROS
HaCaT/ CD4 + T cells/IL-23-induced psoriasis-like mouse model [bib_ref] Anti-inflammatory activity of compounds isolated from Astragalus sinicus L. in cytokine-induced keratinocytes..., Kim [/bib_ref] mechanisms with lower drug doses. Nowadays, transdermal drug delivery of systemic drugs with particular advantages of avoiding first-pass metabolism, lesser side effects, pain-free and noninvasive self-administration for patients brings into investigation [bib_ref] Transdermal drug delivery, Prausnitz [/bib_ref] [bib_ref] Non-invasive delivery strategies for biologics, Anselmo [/bib_ref]. Still, effectively cutaneous drug absorption becomes challenging in the local treatment of psoriasis, particularly for its thickened epidermis [bib_ref] Calderon M. Nanocarriers for skin applications: where do we stand?, Tiwari [/bib_ref].
## Latest developments of biomaterials for psoriasis therapies
For the past few years, numerous studies have explored and optimized more new and refined effective therapeutic modalities for psoriasis with drugs or nanomaterials to circumvent the drawback of conventional drugs and resolve the transdermal approaches limitation of drug diffusion or permeation to the dermis. As a result, switching the dynamic equilibrium of the oxidationreduction system of these key pathogenetic cells is quite pertinent to providing a comprehensive strategy to reshape the immune-microenvironment in psoriasis. Mounting evidence has emphasized the critical role oxidative stress played in the pathogenesis of psoriasis, which promotes the discovery of new therapeutic modalities. Based on the abovementioned reports, ROSmediated dysfunctional different cell types (KCs, skinresident and -infiltrating immune cells functions) in the epithelial microenvironment (EIME) propagate multiple inflammatory loops of psoriasis. Therapies based on ROS-inhibition and -elimination targets for the blockade of inflammatory loops could be effective in the treatment of psoriasis. Besides the systemic and topical antipsoriatic drugs, recent advances in nanotechnology have promoted the emergence of numerous nanosystems, as shown as [fig_ref] Figure 3: Different types of nanoparticles/nanocarriers used as therapeutic modalities of ROS-related psoriasis [/fig_ref] and [fig_ref] Table 3: Nanomaterials used for transdermal drug delivery in psoriasis treatment [/fig_ref] , which could resolve limitations of drug systemic side effects and transdermal drug diffusion or permeation in conventional therapies.
## Self-therapeutic nanomaterials for the treatment of psoriasis mental nanoparticles
Ce-based nanoparticles Ceria nanoparticles (NPs) have been regarded as typical nano-antioxidants with therapeutic effects on a range of ROS-related diseases, including hepatic ischemia-reperfusion injury [bib_ref] Ceria nanoparticles meet hepatic ischemia-reperfusion injury: the perfect imperfection, Ni [/bib_ref] , acute kidney injury (AKI) [bib_ref] Ros-responsive nano-drug delivery system combining mitochondriatargeting ceria nanoparticles with atorvastatin for acute..., Yu [/bib_ref] [bib_ref] Catalytic activity tunable ceria nanoparticles prevent chemotherapy-induced acute kidney injury without interference..., Weng [/bib_ref] , multiple CNS diseases [bib_ref] Custom-made ceria nanoparticles show a neuroprotective effect by modulating phenotypic polarization of..., Zeng [/bib_ref] [bib_ref] Mitochondria-targeting ceria nanoparticles as antioxidants for alzheimer's disease, Kwon [/bib_ref] , rheumatoid arthritis (RA) [bib_ref] Synergistic oxygen generation and reactive oxygen species scavenging by manganese ferrite/ceria co-decorated..., Kim [/bib_ref] , etc. Their detailed mechanism for scavenging the overproduction of ROS from pathogenic cells restores the redox homeostasis for reprogramming the immuno-environment by facilitating the transformation of cytopathogenic phenotypic transition into the cytoprotective subtype. Besides, the ceria NPs could be modified with the capability of localized into mitochondria for reduction of ROS against neuroinflammation [bib_ref] Mitochondria-targeting ceria nanoparticles as antioxidants for alzheimer's disease, Kwon [/bib_ref]. It is well-documented that psoriasis is a disordered oxidative stress-related inflammatory disease, a feasible approach could be manufactured to downregulate oxidative stress for a detoxification effect via direct delivery of ROS-regulating nanosystems into skin lesions. On account of the above ROS-eliminating activity of ceria, it uncovers more opportunities for potential therapeutic interventions to the progress of psoriasis to reconfigure the steady-state cellular redox homeostasis in EIME.
## Wu l. et al. fabricated β-cyclodextrins (β-cds) modified
ceria NPs (β-CDs/CeO 2 NPs) with drug-loaded and antioxidative activities for combinational psoriasis therapy in the IMQ-induced psoriatic model . CeO 2 with intrinsic superoxide dismutase-and catalase-mimicking capacities have been developed as therapeutic agents for cytoprotection against ROS-mediated damageand provides combinational antipsoriatic efficacy for transdermal delivery of dithranol (DIT) [bib_ref] Cyclodextrin-modified ceo2 nanoparticles as a multifunctional nanozyme for combinational therapy of psoriasis, Wu [/bib_ref]. Further research is imperative to broaden better our understanding of the ceria-based NPs and tailor their functional orientations to meet their specific needs for reversing the role of specific redox pathways in the interrelated pathology of psoriasis.
Gold nanoparticles Gold nanoparticles (Au NPs) have shown good biocompatibility, water-solubility, catalytic activity and great potential as self-therapeutic nanosystems for drug delivery platforms against inflammatory disorders, including AKI and RA due to their anti-inflammatory and antioxidative performances [bib_ref] Targeted combination of antioxidative and anti-inflammatory therapy of rheumatoid arthritis using multifunctional..., Li [/bib_ref] [bib_ref] Clinically translatable gold nanozymes with broad spectrum antioxidant and anti-inflammatory activity for..., Zhang [/bib_ref]. It has been reported that the tunable bio-effects of Au NPs differ across research due to the application of regulatory particle sizes and surface modification [bib_ref] Immunomodulatory effects of coated gold nanoparticles in LPS-stimulated and murine model systems, Moyano [/bib_ref]. Özcan A et al.
found that Au NPs, as transdermal drug delivery, could facilitate MTX transcutaneous delivery into the skin across the stratum corneum barriers and lessen psoriatic skin inflammation in noninvasive manners, to avoid Ce NPs-based self-therapeutic nanomaterials for the topical treatment of psoriasis. β-cyclodextrin modified ceria nanoparticles were designed as a ROS scavenger nanozyme to transdermal delivery of dithranol for the combinational therapy of psoriasis. Reproduced with permission [bib_ref] Cyclodextrin-modified ceo2 nanoparticles as a multifunctional nanozyme for combinational therapy of psoriasis, Wu [/bib_ref]. Copyright 2020, Dove Medical Press systemic side effects and achieve good skin penetration, ascribed to small size and immunomodulatory effects of Au NPs [bib_ref] Nanoparticle-coupled topical methotrexate can normalize immune responses and induce tissue remodeling in..., Ozcan [/bib_ref]. Likewise, Au NPs coupled with oligonucleotides (siRNA) can be qualified to preferentially gene editing and enhance the transdermal treatment of psoriasis [bib_ref] Using siRNA-based spherical nucleic acid nanoparticle conjugates for gene regulation in psoriasis, Nemati [/bib_ref]. Additionally, sub-15 nm Au NPs could be tailored by 30% octadecyl chains to restore the deterioration of psoriasis without an excipient and the side effects of hair loss and skin wrinkling [bib_ref] Alkylterminated gold nanoparticles as a self-therapeutic treatment for psoriasis, Han [/bib_ref]. It was attributed to the optimal core size for effective endocytosis by KCs and the assistance of epidermal delivery of Au NPs to effectively restrain the IL-17 signaling pathway mediated the epidermal hyperproliferation and inflammation in the IMQ-induced psoriasis mice model . Therefore, the decisive contributions of these studies in bespoke Au NPs for the intervention of psoriasis make a favorable difference in the biomedical application of Au NPs for the treatment of psoriasis.
## Silver nanoparticles
Recently, considerable research have demonstrated that bio-friendly silver (Ag) NPs have potential properties in immunomodulatory and ROS- Au NPs-based self-therapeutic nanomaterials for the topical treatment of psoriasis. a MTX-GNPs were prepared to inhibit the exacerbation of psoriasis via reshaping the immune infiltration and cytokine secretion of the skin. Reproduced with permission [bib_ref] Skin permeating nanogel for the cutaneous co-delivery of two anti-inflammatory drugs, Shah [/bib_ref]. Copyright 2020, Elsevier. b siRNA conjugated with spherical nucleic acid gold nanoparticles were developed for the reduction of T cell activation and inflammatory gene expression to topically control the progress of psoriasis. Reproduced with permission [bib_ref] Using siRNA-based spherical nucleic acid nanoparticle conjugates for gene regulation in psoriasis, Nemati [/bib_ref]. Copyright 2017, Elsevier. c Alkyl-terminated Au NPs were synthesized as self-therapeutic nanomedicines for topically preventing and treating imiquimod-induced psoriasis mice via downregulation of gene expression involved in the interleukin-17 signaling pathway. Reproduced with permission [bib_ref] Alkylterminated gold nanoparticles as a self-therapeutic treatment for psoriasis, Han [/bib_ref]. Copyright 2017, American Chemical Society modulating activities by elaborately tailoring their size and shape [bib_ref] The impact of engineered silver nanomaterials on the immune system, Ninan [/bib_ref] [bib_ref] Targeted silver nanoparticles for rheumatoid arthritis therapy via macrophage apoptosis and re-polarization, Yang [/bib_ref]. AgNPs decorate biomaterials with appropriately therapeutic window concentrations of Ag + ions, not only can they endow AgNPs with the biological function of regulating macrophage polarization and ROS responsiveness but also optimize their biocompatibility for alleviating a wide variety of preclinical inflammatory diseases such as RA and diabetic wound [bib_ref] The impact of engineered silver nanomaterials on the immune system, Ninan [/bib_ref] [bib_ref] Targeted silver nanoparticles for rheumatoid arthritis therapy via macrophage apoptosis and re-polarization, Yang [/bib_ref] [bib_ref] Improved immunoregulation of ultra-low-dose silver nanoparticle-loaded TiO2 nanotubes via M2 macrophage polarization..., Chen [/bib_ref] [bib_ref] Bergenin loaded gum xanthan stabilized silver nanoparticles suppress synovial inflammation through modulation..., Rao [/bib_ref] [bib_ref] Silver nanoparticles: advanced and promising technology in diabetic wound therapy, Choudhury [/bib_ref]. Ag NPs extracted from natural herbs efficiently suppressed NF-κB activation of macrophage in vitro and human psoriasis plaques, eventually resulting in psoriasis resolution [bib_ref] Topical silver and gold nanoparticles complexed with cornus mas suppress inflammation in..., Crisan [/bib_ref]. Furthermore, immunomodulatory Ag NPs co-decorated ZnO nanoparticles were conferred with the capability of inactivating p65 in proinflammatory macrophages and abrogating the secretion of ROS-induced adaptive cytokines in psoriatic KCs . These composite nanoparticles (Ag/ZnO NPs) identified as self-therapeutic nanocarriers to deliver MTX into the stratum corneum, not only exerted their immunosuppressive effect but also combinedly augment the antipsoriatic efficacy of a lowdose MTX under the realization of sustainable MTX release [bib_ref] A multifunctional composite hydrogel as an intrinsic and extrinsic coregulator for enhanced..., Xu [/bib_ref]. Therefore, these results suggested that the appropriate concentration of Ag NPs could be designed for anti-inflammation and ROS-depletion against inflammatory disorders.
## Polymers
It is worth mentioning that multifarious polymers with different modifications are available for a wide range of biomedical applications, including drug delivery systems [bib_ref] Surfactant-free, self-assembled nanomicelles-based transdermal hydrogel for safe and targeted delivery of methotrexate..., Qindeel [/bib_ref] , gene targeting [bib_ref] Polymer-stabilized Cas9 nanoparticles and modified repair templates increase genome editing efficiency, Nguyen [/bib_ref] [bib_ref] Nanoparticle delivery of Cas9 ribonucleoprotein and donor DNA in vivo induces homology-directed..., Lee [/bib_ref] , and therapeutic agents [bib_ref] Tuned cationic dendronized polymer: molecular scavenger for rheumatoid arthritis treatment, Peng [/bib_ref] [bib_ref] Cationic nanoparticle as an inhibitor of cell-free DNAinduced inflammation, Liang [/bib_ref] for targeted therapy in inflammatory diseases. Cell-free DNA (cfDNA) has been proven to be an inflammatory trigger to activate DNA sensors-induced immune responses involved in initiating and exacerbating the pathogenesis of autoimmune diseases [bib_ref] Circulating cell-free DNA levels in portuguese patients with psoriasis vulgaris according to..., Coimbra [/bib_ref] [bib_ref] Circulating free DNA and its emerging role in autoimmune diseases, Mondelo-Macía [/bib_ref] , such as RA, SLE and psoriasis. It presents evidence that approaches for effectively eliminating cfDNA is feasible for the remission of disease severity. Liang H et al. constructed self-assembly of PLGA-block-PDMA block copolymer, PLGA-b-PDMA 463 with a high DNA-binding affinity, which could scavenge cfDNA released from dead and dying cells to restrain autoimmune inflammation against RA [bib_ref] Cationic nanoparticle as an inhibitor of cell-free DNAinduced inflammation, Liang [/bib_ref]. In psoriasis, these cationic nanoparticles were composed of the diblock copolymer of PLGAb-PDMA 474 , which similarly beneficially prevented cfDNA from the formation of the DNA-LL37 immune Ag NPs-based self-therapeutic nanomaterials for the topical treatment of psoriasis. The Car@NMs@MTX-ZA hydrogel was successfully fabricated as self-therapeutic nanotherapy for combined anti-inflammation with antiproliferation for the treatment of psoriasis. Reproduced with permission [bib_ref] A multifunctional composite hydrogel as an intrinsic and extrinsic coregulator for enhanced..., Xu [/bib_ref]. Copyright 2022, Springer Nature complex via topical application against psoriasis [bib_ref] Topical nanoparticles interfering with the DNA-LL37 complex to alleviate psoriatic inflammation in..., Liang [/bib_ref]. Altogether, these data implied that the possible applications of bespoke polymers could neutralize the detrimental effects of cfDNA or RNA signature to serve as potential antipsoriatic nanomedicines.
## Natural bioactive compound
Natural products have gained considerable attention for psoriasis treatment due to their excellent biocompatibility and high effectiveness. Bilirubin, a highly potent anticancer and anti-inflammatory compound can scavenge various ROS and plays a crucial role in protecting cells from oxidative stress-mediated damage in the human body [bib_ref] Bilirubin nanoparticles as a nanomedicine for anti-inflammation therapy, Lee [/bib_ref]. [bib_ref] Bilirubin nanomedicine alleviates psoriatic skin inflammation by reducing oxidative stress and suppressing..., Keum [/bib_ref]. Polyphenols and flavonoids in natural products have been widely used in the treatment of inflammation-related diseases due to their excellent antioxidative properties. Recently, mung bean-derived NPs (MBNs) with a facile approach has been reported for alleviating skin inflammation. MBNs can regulate macrophage polarization and antagonize the activation of the nuclear factor kappa B (NF-κB) signaling pathway which are conducive to the subsides of inflammation in psoriasiform skin [bib_ref] Transcutaneous delivery of mung bean-derived nanoparticles for amelioration of psoriasis-like skin inflammation, Sun [/bib_ref]. Moreover, melatonin (MLT), a natural hormone and antioxidant mainly derived from the pineal gland with the circadian rhythm of secretion, have regarded as an anti-inflammation and immunomodulator for inflammatory skin diseases [bib_ref] Biohybrid hydrogel comprising collagen-capped silver nanoparticles and melatonin for accelerated tissue regeneration..., Ragothaman [/bib_ref] [bib_ref] Melatonin's impact on antioxidative and antiinflammatory reprogramming in homeostasis and disease, Chitimus [/bib_ref] [bib_ref] On the role of melatonin in skin physiology and pathology, Slominski [/bib_ref] [bib_ref] Effect of melatonin on psoriatic phenotype in human reconstructed skin model, Scuderi [/bib_ref] , such as skin psoriasis [bib_ref] Effect of melatonin on psoriatic phenotype in human reconstructed skin model, Scuderi [/bib_ref] and wound healing [bib_ref] Melatonin loaded lecithin-chitosan nanoparticles improved the wound healing in diabetic rats, Lopes Rocha Correa [/bib_ref]. Several studies have shown that the circadian rhythm of melatonin secretion in psoriatic patients is disappeared and melatonin-dependent redox homeostasis of the skin cells is dysregulated [bib_ref] Effect of melatonin on psoriatic phenotype in human reconstructed skin model, Scuderi [/bib_ref] [bib_ref] Melatonin, mitochondria, and the skin, Slominski [/bib_ref]. Topical or systemic administration of melatonin could make good effective in diminishing the extensive ROS generation and proinflammatory cytokines under psoriasis and skin tissue regeneration [bib_ref] Biohybrid hydrogel comprising collagen-capped silver nanoparticles and melatonin for accelerated tissue regeneration..., Ragothaman [/bib_ref] [bib_ref] Effect of melatonin on psoriatic phenotype in human reconstructed skin model, Scuderi [/bib_ref]. Taken together, these biologically-derived antioxidant NPs have not only significant efficacy but also high clinical translation potential.
## Nanomaterial-based transdermal drug delivery platform for the treatment of psoriasis
Other than the aforesaid representatively self-therapeutic nanoparticles for the topical restoration of psoriasis.
Recently, several nanocarriers, such as liposomes [bib_ref] Oleic acid as the active agent and lipid matrix in cilomilast-loaded nanocarriers..., Lin [/bib_ref] [bib_ref] Nanoparticle-assisted transcutaneous delivery of a signal transducer and activator of transcription 3-inhibiting..., Kim [/bib_ref] , polymers [bib_ref] Topical cationic hairy particles targeting cell free DNA in dermis enhance treatment..., Yan [/bib_ref] [bib_ref] Topical nanoparticles interfering with the DNA-LL37 complex to alleviate psoriatic inflammation in..., Liang [/bib_ref] , silica nanoparticles [bib_ref] Topical cationic hairy particles targeting cell free DNA in dermis enhance treatment..., Yan [/bib_ref] [bib_ref] Development of erianin-loaded dendritic mesoporous silica nanospheres with pro-apoptotic effects and enhanced..., Mo [/bib_ref] , metal nanoparticles [bib_ref] Nanoparticle-coupled topical methotrexate can normalize immune responses and induce tissue remodeling in..., Ozcan [/bib_ref] [bib_ref] Alkylterminated gold nanoparticles as a self-therapeutic treatment for psoriasis, Han [/bib_ref] and microneedles [bib_ref] Hyaluronic acid-based dissolving microneedle patch loaded with methotrexate for improved treatment of..., Du [/bib_ref] have been introduced to favor transdermal delivery of antipsoriatic drugs and gene editing efficiency, which Polymer-based self-therapeutic nanomaterials for the topical treatment of psoriasis. Cationic nanoparticles were constructed as cfDNA scavengers for topical remission of DNA-LL37-induced cell inflammation in a psoriasiform mice model and cynomolgus monkey model. Reproduced with permission [bib_ref] Topical nanoparticles interfering with the DNA-LL37 complex to alleviate psoriatic inflammation in..., Liang [/bib_ref]. Copyright 2020, American Association for the Advancement of Science strategically make contributions to avoidance of their low solubility, bioavailability, and poor skin permeability to augment their antipsoriatic efficacy.
## Lipid nanoparticles
It is widely recognized that lipid nanoparticles have been widely used in skin-related diseases [bib_ref] Nanoparticle-assisted transcutaneous delivery of a signal transducer and activator of transcription 3-inhibiting..., Kim [/bib_ref] [bib_ref] Bragazzi NL. Nanodermatology-based solutions for psoriasis: state-of-the art and future prospects, Damiani [/bib_ref] and skinbased cosmetics [bib_ref] Application of nanoparticles in percutaneous delivery of active ingredients in cosmetic preparations, Khezri [/bib_ref] , owing to their excellent bioavailability and biodegradability. Their comprehensive roles of both topical drug carriers and penetration enhancers, improve transdermal delivery of drugs [bib_ref] Oleic acid as the active agent and lipid matrix in cilomilast-loaded nanocarriers..., Lin [/bib_ref] [bib_ref] Curcumin-loaded lipid-hybridized cellulose nanofiber film ameliorates imiquimod-induced psoriasis-like dermatitis in mice, Kang [/bib_ref] [bib_ref] Enhanced transdermal efficiency of curcumin-loaded peptide-modified liposomes for highly effective antipsoriatic therapy, Yu [/bib_ref] , peptides [bib_ref] Nanoparticle-assisted transcutaneous delivery of a signal transducer and activator of transcription 3-inhibiting..., Kim [/bib_ref] , and oligonucleotide [bib_ref] Targeting the IL-17 receptor using liposomal spherical nucleic acids as topical therapy..., Liu [/bib_ref] [bib_ref] TNFα siRNA delivery by nanoparticles and photochemical internalization for psoriasis topical therapy, Suzuki [/bib_ref] into skin lesions. Kim JY et al. designed STAT3-inhibiting peptide-encased discoidal lipid nanoparticles (DLNPs) that could contribute to promoting the penetration of peptide inhibitors into thicked stratum corneum of psoriasis . Meanwhile, these lipid formulation-based transcutaneous delivery systems exerted good biocompatibility without the side effects of conventional corticosteroid drugs [bib_ref] Nanoparticle-assisted transcutaneous delivery of a signal transducer and activator of transcription 3-inhibiting..., Kim [/bib_ref]. In addition, Suzuki IL et al. fabricated polymer-lipid nanoparticles (PLNs) to resolve the delivery limitation of RNAi topical therapy, such as improving the biological stability of siRNA, optimizing its cellular endocytosis and sufficient endosomal release Lipid nanomaterials-based transdermal drug delivery platform for the treatment of psoriasis. a The preparation of the DLNP transcutaneous delivery system could improve the skin penetration of STAT3-inhibiting peptides for efficiently treating psoriatic skin inflammation without causing adverse systemic events. Reproduced with permission [bib_ref] Nanoparticle-assisted transcutaneous delivery of a signal transducer and activator of transcription 3-inhibiting..., Kim [/bib_ref]. Copyright 2018, American Chemical Society. b Hybrid polymer-lipid nanoparticles in combinational with photosensitizer TPPS2a for delivery of siRNA were aimed to topical treat psoriasis effectively through optimizing the endosomal escape of TNFα siRNA in the cytoplasm. Reproduced with permission [bib_ref] TNFα siRNA delivery by nanoparticles and photochemical internalization for psoriasis topical therapy, Suzuki [/bib_ref]. Copyright 2021, Elsevier. c Lipid-hybridized CNF film was successfully prepared for transdermal delivery of curcumin to cure psoriasis. Reproduced with permission [bib_ref] Curcumin-loaded lipid-hybridized cellulose nanofiber film ameliorates imiquimod-induced psoriasis-like dermatitis in mice, Kang [/bib_ref]. Copyright 2018, Elsevier [bib_ref] TNFα siRNA delivery by nanoparticles and photochemical internalization for psoriasis topical therapy, Suzuki [/bib_ref]. Analogously, curcumin-loaded cellulose nanofiber (CNF) films composed of hybridized curcumin (Cur)loaded nanostructured lipid carriers (NLCs) were constructed to enhance the deposition of curcumin into the dermis via topical treatment, conducing to amelioration of the psoriatic skin symptoms in IMQ-induced mice, almost comparable to topical corticosteroid cream [bib_ref] Curcumin-loaded lipid-hybridized cellulose nanofiber film ameliorates imiquimod-induced psoriasis-like dermatitis in mice, Kang [/bib_ref].
Another report also demonstrated that curcumin-loaded hyaluronan (HA)-modified ethosomes could target overexpressed CD44 protein and allowed the slow release of the loaded curcumin in the inflamed epidermis [bib_ref] CD44 assists the topical anti-psoriatic efficacy of curcumin-loaded hyaluronan-modified ethosomes: a new..., Zhang [/bib_ref]. Yet the limitation of lipid nanoparticles is vulnerable to oxidative degradation and exhibits poor stability, resulting in lower drug payload and inconvenient storage. These carrier systems may not have the capacity of prolonging circulation and retention, leading to a limit in the systemic bioavailability and therapeutic efficacy of cargos. More efforts should be made to optimize the facility of lipid nanoparticles.
## Silica nanoparticles
It is well-demonstrated that mesoporous silica nanoparticles have been considered as available drug/gene delivery carriers for their unique properties and biocompatibility. They could be functionalized with specific properties via tuning their size and surface modification/ bioconjugation for targeting and delivering therapeutic agents against a variety of inflammatory diseases [bib_ref] Mesoporous silica nanoparticles: synthesis, biocompatibility and drug delivery, Tang [/bib_ref] , such as RA [bib_ref] Synergistic oxygen generation and reactive oxygen species scavenging by manganese ferrite/ceria co-decorated..., Kim [/bib_ref] , osteoporosis [bib_ref] Osteoporosis remission and new bone formation with mesoporous silica nanoparticles, Mora-Raimundo [/bib_ref] , and atherosclerosis [bib_ref] Targeting and clearance of senescent foamy macrophages and senescent endothelial cells by..., Pham [/bib_ref] , etc. Owing to the abovementioned advantages of silica NPs, Mo C et al. employed dendritic mesoporous silica NPs as drug carriers to enhance the penetration activity of erianin across the skin in the favor of exerting an inhibitory effect on keratinocyte proliferation for the topical therapy of psoriasis [bib_ref] Development of erianin-loaded dendritic mesoporous silica nanospheres with pro-apoptotic effects and enhanced..., Mo [/bib_ref]. Moreover, the skin retention and permeability of silica NPs could be regulated by the particle size and polymer decoration, thereby affecting their affinity to cfDNA in the dermis along with regulation of the antipsoriatic effects [bib_ref] Topical cationic hairy particles targeting cell free DNA in dermis enhance treatment..., Yan [/bib_ref]. As a result of these positive results, it is encouraging that the versatile well-controlled and -modified fabrication of silica NPs has great potential to clinically apply to treat cutaneous inflammatory diseases.
## Polymer/nanomicelles
It is widely known that polymer/nanomicelles can promote targeted therapy and sustained hydrophobic drug delivery with relatively high drug loading capacity, except for their performance as cfDNA scavengers. Because of their capability of prolonged circulation, reducing the initial-burst release and delivery of the hydrophobic drug, they are often utilized as a carrier system for transdermal drug delivery to resolve the restriction of drug controlled release and percutaneous absorption, thereby circumventing the drug-associated side effects [bib_ref] Multi-component clobetasol-loaded monolithic lipid-polymer hybrid nanoparticles ameliorate imiquimod-induced psoriasis-like skin inflammation in..., Pukale [/bib_ref] [bib_ref] A multifunctional composite hydrogel as an intrinsic and extrinsic coregulator for enhanced..., Xu [/bib_ref] [bib_ref] Surfactant-free, self-assembled nanomicelles-based transdermal hydrogel for safe and targeted delivery of methotrexate..., Qindeel [/bib_ref]. Polycaprolactone-Polyethyleneglycol-Polycaprolactone (PCL-PEG-PCL)-based self-assembled nanomicelles were employed as a carrier system for efficient delivery and sustainable release of MTX against RA and psoriasis through the transdermal route [bib_ref] A multifunctional composite hydrogel as an intrinsic and extrinsic coregulator for enhanced..., Xu [/bib_ref] [bib_ref] Surfactant-free, self-assembled nanomicelles-based transdermal hydrogel for safe and targeted delivery of methotrexate..., Qindeel [/bib_ref]. Similarly, the stable multi-component monolithic lipid-polymer hybrid nanoparticles (LPNs) were fabricated to load clobetasol propionate, a potent corticosteroid, contributing to facilitating its sustained release and penetration into deeper dermis, consequently exhibiting enhanced therapeutic effect at dose reduction without systemic toxicities absorption of the corticosteroids [bib_ref] Multi-component clobetasol-loaded monolithic lipid-polymer hybrid nanoparticles ameliorate imiquimod-induced psoriasis-like skin inflammation in..., Pukale [/bib_ref]. However, the therapeutic efficacy of topical administration is compromised by the comprehensive effect of limited penetration and skin retention. Yang Mai et al. developed the tris (hydroxymethyl) aminomethane-modified bioadhesive nanoparticles (Tris-BNPs) encapsulated with betamethasone dipropionate (BD) which showed deeper penetration and longer retention compared with commercial BD ointment. This formulation can mitigate skin inflammation after only a single administration [bib_ref] Topical formulation based on disease-specific nanoparticles for single-dose cure of psoriasis, Mai [/bib_ref]. Thus, all these present works demonstrated polymers with good drug loading capacity, biocompatibility, stability, drug controlled release and efficient cellular uptake, possessed great potential for pharmaceutical applications in the field of transdermal drug delivery systems. However, the drug capacity strongly depends on the concentrations of nanomicelles [bib_ref] Targeted therapy in chronic diseases using nanomaterial-based drug delivery vehicles, Singh [/bib_ref]. Strategies should be innovated to combine the advantages of different nanoparticles to achieve most of the benefits of improved transcutaneous antipsoriatic efficacy.
## Microneedles
Emerging nanotechnologies based-microneedles associated with efficient settlement for the dilemma of skin penetration hold tremendous promise in transdermal delivery therapy [bib_ref] Transdermal drug delivery, Prausnitz [/bib_ref] [bib_ref] Microneedles-based drug delivery for the treatment of psoriasis, Shravanth [/bib_ref]. Microneedles are capable of traversing the stratum corneum in a micro-invasive manner and directly translocating bioactive drugs into the dermis [bib_ref] Hyaluronic acid-based dissolving microneedle patch loaded with methotrexate for improved treatment of..., Du [/bib_ref] [bib_ref] Microneedle-assisted genome editing: a transdermal strategy of targeting by CRISPR-Cas9 for synergistic..., Wan [/bib_ref] [bib_ref] Polymeric microneedles for transdermal protein delivery, Ye [/bib_ref] [bib_ref] Keratinocyte membrane-mediated nanodelivery system with dissolving microneedles for targeted therapy of skin..., Jing [/bib_ref]. It could be equipped with various therapeutic efficacies via the incorporation of appropriate structural nanomaterials, genome editing materials as well as drug molecules or nanomedicines with tailored pharmacological properties. Wan T et al. had taken advantage of the CRISPR-Cas9-based genome editing technology for precisely targeting the inflammatory signatures of NLRP3, which mediated abnormal crosstalking of innate and adaptive immune responses and glucocorticoid resistance in psoriasis [bib_ref] Microneedle-assisted genome editing: a transdermal strategy of targeting by CRISPR-Cas9 for synergistic..., Wan [/bib_ref]. More importantly, the presence of a microneedles-mediated transdermal therapeutic strategy positively reduced off-target effects of gene editing by allowing the local release of genome editor in target lesions of psoriasis and atopic dermatitis to improve glucocorticoid sensitivity [fig_ref] Figure 11: Microneedles-based transdermal drug delivery platform for the treatment of psoriasis [/fig_ref]. Additionally, Q. Jing et al. utilized the homologous targeting functions of the HaCaT cell membrane to develop HaCaT cell membrane-coated nanocarriers for Silica nanomaterials-based transdermal drug delivery platform for the treatment of psoriasis. a The synthesis of erianin-loaded dendritic mesoporous silica was employed for topical therapy of psoriasis, ascribed for their mechanisms on pro-apoptotic effect in KCs. Reproduced with permission [bib_ref] Development of erianin-loaded dendritic mesoporous silica nanospheres with pro-apoptotic effects and enhanced..., Mo [/bib_ref]. Copyright 2020, Springer Nature. b Optimized size of silica NPs decorated with polymer could elevate the affinity of cfDNA to inhibit topical psoriasis inflammation via better penetration ability. Reproduced with permission [bib_ref] Topical cationic hairy particles targeting cell free DNA in dermis enhance treatment..., Yan [/bib_ref]. Copyright 2021, Elsevier transdermal targeted delivery of shikonin in the pathological epidermis, as shown in [fig_ref] Figure 11: Microneedles-based transdermal drug delivery platform for the treatment of psoriasis [/fig_ref] -d. This nanocomposite could be internalized by the KCs, leading to the triggering of drug release in the target lesion. Ultimately, the augmented therapeutic efficacy of shikonin against imiquimod-induced psoriatic epidermal hyperplasia was achieved [bib_ref] Keratinocyte membrane-mediated nanodelivery system with dissolving microneedles for targeted therapy of skin..., Jing [/bib_ref]. As surveyed above, whereas therapeutic drug delivery through microneedles, has received considerable attention for different applications in the field of dermatology, the potential skin bacterial, fungal infection-associated risks, sensitization, and other restrictions of the costs, transportation, cargoes stability, and loading are inevitable [bib_ref] Polymeric microneedles for transdermal protein delivery, Ye [/bib_ref]. More studies should be investigated to optimize the biocompatibility of microneedles before being applied to human skin. Meanwhile, further schemes of ingredients should be facilitated to resolve the above limitation and optimize the clinical translation of formulations.
## Hydrogel
In consideration of multiple inflammatory pathways of psoriasis immunopathogenesis and optimization of topical drug bioavailability, inhibition of psoriasis activity with multiple therapeutic modalities specific to different targets outbalance single-agent approaches. Consequently, an ideal percutaneous nanocarrier needs to meet the following requirements: (1) self-therapeutic activity, with intrinsic anti-inflammatory property and improved therapeutic efficacy of extrinsic medication; (2) better drug loading capacity and controllable drug release; (3) good moisture retention, which can maintain the moist environment of the skin and reduction of drug breakage; and (4) enhanced patient compliance. Hydrogels, owing to their biochemical characteristics of good retention, avoidance of drug leakage, good hydrophilicity and adhesiveness, have been identified as the most competitive candidate for the percutaneous treatment of inflammatory diseases [bib_ref] Functional hydrogels as wound dressing to enhance wound healing, Liang [/bib_ref] [bib_ref] Thermoresponsive in situ forming hydrogel with sol-gel irreversibility for effective methicillin-resistant staphylococcus..., Yan [/bib_ref] [bib_ref] Plant-inspired adhesive and tough hydrogel based on ag-lignin nanoparticles-triggered dynamic redox catechol..., Gan [/bib_ref]. Considerable research has demonstrated that hydrogels can be well-appointed with tunable functions via the incorporation of various bioactive substances, such as nanoparticles and drugs and establish well-pleasing biomedical applications in transdermal drug delivery [bib_ref] Surfactant-free, self-assembled nanomicelles-based transdermal hydrogel for safe and targeted delivery of methotrexate..., Qindeel [/bib_ref] [bib_ref] Rational design of immunomodulatory hydrogels for chronic wound healing, Kharaziha [/bib_ref] [bib_ref] A biodegradable hydrogel system containing curcumin encapsulated in micelles for cutaneous wound..., Gong [/bib_ref] [bib_ref] Enhanced topical penetration, system exposure and anti-psoriasis activity of two particle-sized, curcumin-loaded..., Sun [/bib_ref] [bib_ref] Hydrogel-mediated topical delivery of steroids can effectively alleviate psoriasis attenuating the autoimmune..., Rana [/bib_ref]. As shown in [fig_ref] Figure 2: ROS contributes to the rearranging immunometablism of different cell types, accompanied by... [/fig_ref] , For improvement of the transdermal application of lyophobic drugs, Sun L et al. fabricated curcumin (Cur) loaded poly (lactic-co-glycolic acid) (PLGA) nanoparticles (NPs) loaded into the hydrogel which was employed to topically treat IMQ-induced psoriasis-like mouse for promotion of drug permeability across skin and enhancement of anti-psoriatic activity [fig_ref] Figure 2: ROS contributes to the rearranging immunometablism of different cell types, accompanied by... [/fig_ref] [bib_ref] Enhanced topical penetration, system exposure and anti-psoriasis activity of two particle-sized, curcumin-loaded..., Sun [/bib_ref]. similarly, Qiu F et al. produced Celastrol Noisome hydrogel (Cel Nio gel) for topical administration to psoriasis. When applied in the IMQ-induced psoriatic mice model, cel was mainly accumulated in the skin other than exposure Polymer/nanomicelles-based transdermal drug delivery platform for the treatment of psoriasis. Lipid-polymer hybrid nanoparticles were fabricated to load clobetasol propionate for enhancement of its cellular uptake and skin permeability to improve antipsoriatic efficacy. Reproduced with permission [bib_ref] Multi-component clobetasol-loaded monolithic lipid-polymer hybrid nanoparticles ameliorate imiquimod-induced psoriasis-like skin inflammation in..., Pukale [/bib_ref]. Copyright 2020, Elsevier Hydrogel-based transdermal drug delivery platform for the treatment of psoriasis. a Cur encapsulated into PLGA NPs were synthesized as hydrogel to optimize the dispersion, sustained release and penetration of curcumin across the skin for improvement of its anti-psoriatic efficacy. Reproduced with permission [bib_ref] Enhanced topical penetration, system exposure and anti-psoriasis activity of two particle-sized, curcumin-loaded..., Sun [/bib_ref]. Copyright 2017, Elsevier. b Therapeutic mechanism of Cel Nio gel for the transcutaneous treatment of imiquimod-induced psoriasiform skin inflammation. Reproduced with permission [bib_ref] Celastrol niosome hydrogel has anti-inflammatory effect on skin keratinocytes and circulation without..., Qiu [/bib_ref]. Copyright 2021, Dove Medical Press to the blood or lymphatic system, resulting in the reduction of the mRNA levels of inflammatory cytokines [fig_ref] Figure 2: ROS contributes to the rearranging immunometablism of different cell types, accompanied by... [/fig_ref] [bib_ref] Celastrol niosome hydrogel has anti-inflammatory effect on skin keratinocytes and circulation without..., Qiu [/bib_ref]. Additionally, Kajal Rana et al. presented that a betamethasone-loaded topical hydrogel (B-Gel) which can efficiently entrap steroids with the properties of spreadability and sustained release drugs, provided an alternative for topical application of steroids [bib_ref] Hydrogel-mediated topical delivery of steroids can effectively alleviate psoriasis attenuating the autoimmune..., Rana [/bib_ref]. Moreover, implementing biocompatible hydrogel micropatch probes integrated with mass spectrometry to explore the skin metabolome could be regarded as a diagnostic approach to provide information about the pathological alterations of the skin metabolome caused by psoriasis, favoring understanding of the complicated pathophysiology. However, antibiotic-immobilized hydrogels should be seriously utilized due to the problems of multidrug resistance and relatively long treatment course, while hydrogels loaded with noble metal nanoparticles often cause undesirable systemic toxicity. Above all, It is noted that most of the existing ROS-based nanomedicines or transdermal delivery nanoplatform are engineered with some deficiency, comprehensive resolution of limitations of these nanobiotechnological carriers related to drug controlled release, drug lower loading capacity and optimizing transdermal permeation, particularly in the thickened stratum corneum of psoriasis remains intractable. Therefore, it is highly expected that address these issues in elaborately engineered redox-active nanosystems design and a more simplified way for the feasibility of clinical translation, rather than decorating sophisticated structures that may render potential biosafety issues.
## Summary and outlook
As the significant role of oxidative stress in the molecular pathological mechanisms of psoriasis continues to be unraveled, targeting ROS in dysfunctional different cell types in EIME offers a promising methodology for psoriasis. In the future, a more major focus should be paid to investigating more effectively new-generation of therapeutics mediated precisely regulation of cellular ROS concentrations in EIEM within a physiological threshold. Meanwhile, it is appreciated that the noticeable advances in the field of nanotechnology regarding multifarious nanomaterials with ROS depletion performances have been witnessed. Most notably, besides current ROS-detoxifying self-therapeutic nanomaterials directly against psoriasis, the emergence of a nano-platform for transdermal drug delivery system greatly expands the application of nanomaterials in the field of precision medicine. Nanotechnologies dramatically facilitate the absorption and diffusion of drugs at skin barriers, especially in psoriatic conditions characterized by highly packed SC, giving rise to increased drug availability in local therapy and decreased systemic adverse effects. The incorporation of nanotechnologies offers protection for the labile therapeutically active compounds as well as the assistance of drug storage and prolonged residence time of drug molecules at the target site against skin disease. Aside from the mentioned already, it is anticipated that more comprehensive investigations related to reconstructed skin experimental models should mimic the realtime biological status of skin lesions for the achievement of accessing the permeability and pharmaceutical properties of nanomaterials. Furthermore, the skin irritation and biosafety evaluations of nanomaterials about long-term therapeutic effects should be conducted for potential clinical transformation. Finally, we envision that these nanobiotechnologies will expand more therapeutic avenues for precision medicine, especially in skin diseases.
[fig] Figure 2: ROS contributes to the rearranging immunometablism of different cell types, accompanied by exerting their effector functions in response to tissue environments via intermediating the main cellular oxidation-reduction (redox) signaling pathways and immune-resident or -infiltrating cells under psoriasis conditions. [/fig]
[fig] Figure 3: Different types of nanoparticles/nanocarriers used as therapeutic modalities of ROS-related psoriasis [/fig]
[fig] Figure 11: Microneedles-based transdermal drug delivery platform for the treatment of psoriasis. a Microneedle-mediated transdermal codelivery of CRISPR-Cas9-based genome editor and glucocorticoids were used for high-efficiency treatment of psoriasis. Reproduced with permission[168].Copyright 2021, American Association for the Advancement of Science. b Characterization images of the MN patches, CP/Ad-SS-GD/Cas9 RNP nanoparticles and Dex-loaded PLGA nanoparticles; drug release of Cas9 protein and Dex from the MN patch; fluorescence images of MN patch. Reproduced with permission [168]. Copyright 2021, American Association for the Advancement of Science. c Schematic illustration of the synthesis of SKN-PMs and HCM/SKN-PMs. Reproduced with permission [216]. Copyright 2021, Elsevier. d Sketch of the MN-HCM/SKN-PM preparation process and their characterization images. Reproduced with permission [216]. Copyright 2021, Elsevier Fig. 12 [/fig]
[table] Table 3: Nanomaterials used for transdermal drug delivery in psoriasis treatment [/table]
|
Subsets and Splits